Backtestslitellmrun-20261111-220245#41394
[Bug]: All chatgpt/* models missing reasoning annotations present on their openai/ and azure/ twins
Our patch was judged better than the merged fix.
In this judgment, Patch B = our agent, Patch A = the merged fix.
Patch B resolves the full ask: all 14 chatgpt/* rows get supports_reasoning plus the effort flags of their openai twins, and chatgpt/gpt-5.6-{sol,terra,luna} get supports_max_reasoning_effort as the issue requested. The independent tests confirm this: B passes every scenario including the two rows without openai twins (gpt-5.3-codex-spark, gpt-5.3-instant) and the whole-map jq-style check. Patch A leaves those two rows unannotated (three independent tests still FAIL-ASSERT on A) and never adds supports_max_reasoning_effort anywhere, so the secondary ask is unaddressed. A also bundles a large amount of unrelated churn: repriced azure gpt-5.6 and bing_grounding entries, new anthropic batch pricing, new azure_ai/bedrock_mantle/xai/nebius models, Rust ModelInfo struct fields, schema and UI type changes, and changes to existing test expectations. That churn adds real risk (pricing changes affect billing) and obscures the fix. On tests, B adds parity checks against both the main and backup maps, a resolve_supported_reasoning_efforts check for the max tier, and a get_model_info check mirroring the issue's /model/info scenario; A's single parity test is reasonable but scoped only to openai twins, which is exactly why it misses the two rows A left broken. One small gap in B: it omits default_reasoning_effort on chatgpt/gpt-5.4 and gpt-5.2 where the openai twins carry it, and B's test does not check that key; A does carry it over. That is minor relative to A's incomplete coverage and unrelated scope. B is the better fix.
Four judges ruled on this ticket in parallel, and the verdict comes from counting their votes. Each saw the issue and the two patches, labelled A and B in random order (two of them saw ours as A), with issue links and version numbers removed. Where tests could check the fix, they also saw the results of tests a separate agent wrote for the issue, run on both patches. None was told which patch was ours. The reasoning above is one judge's, shown verbatim.
Solve
32m 31s
Wall-clock time in the container
Full test suite
No new failures
11,546 passed · 44 skipped · 11,596 total
Contamination audit
Clean
No sign the agent saw the real fix
Patch size
204vs 1,166 merged
Changed lines, ours vs merged
The issue
View on GitHubWhat happened
Every model entry under the chatgpt/ provider prefix is missing its reasoning-capability annotations, even though the identical underlying models are fully annotated under the openai/ and azure/ prefixes.
All 14 chatgpt/* entries in model_prices_and_context_window.json (main) have zero reasoning keys — no supports_reasoning, no supports_*_reasoning_effort:
$ jq -r '[to_entries[]|select(.key|startswith("chatgpt/"))] |
"total: \(length)",
"with supports_reasoning: \([.[]|select(.value.supports_reasoning != null)]|length)",
"with any effort flag: \([.[]|select(.value|to_entries|any(.key|test("reasoning_effort")))]|length)"' \
model_prices_and_context_window.json
total: 14
with supports_reasoning: 0
with any effort flag: 0
Affected keys:
chatgpt/gpt-5.1-codex-max chatgpt/gpt-5.3-codex-spark chatgpt/gpt-5.6-luna
chatgpt/gpt-5.1-codex-mini chatgpt/gpt-5.3-instant chatgpt/gpt-5.6-sol
chatgpt/gpt-5.2 chatgpt/gpt-5.4 chatgpt/gpt-5.6-terra
chatgpt/gpt-5.2-codex chatgpt/gpt-5.4-pro
chatgpt/gpt-5.3-chat-latest chatgpt/gpt-5.5
chatgpt/gpt-5.3-codex
The inconsistency
Same model, three prefixes, two different answers:
// chatgpt/gpt-5.6-sol — no reasoning keys at all
{
"litellm_provider": "chatgpt",
"source": "https://platform.openai.com/docs/models/gpt-5.6-sol",
"max_input_tokens": 1050000,
"max_output_tokens": 128000,
"mode": "responses",
"supported_endpoints": ["/v1/chat/completions", "/v1/responses"],
"supports_function_calling": true,
"supports_vision": true
}
// gpt-5.6-sol (litellm_provider: openai) — annotated
{
"mode": "chat",
"supports_reasoning": true,
"supports_none_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"supports_xhigh_reasoning_effort": true
}
// azure/gpt-5.6-sol — annotated
{
"mode": "chat",
"supports_reasoning": true,
"supports_none_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"supports_xhigh_reasoning_effort": true
}
chatgpt/gpt-5.6-sol is a reasoning model — it accepts reasoning_effort, and /model/info lists it in supported_openai_params. It just never says so in the cost map.
Impact
Any consumer that reads capabilities from /model/info sees a reasoning model advertising no reasoning support. Concretely, for a route declared as:
model_list:
- model_name: high
litellm_params:
model: chatgpt/gpt-5.6-sol
/model/info returns supports_reasoning: null and every supports_*_reasoning_effort: null. Downstream clients that gate their thinking/effort UI on these flags disable reasoning controls entirely, or fall back to a hardcoded subset. The only workaround is to hand-write model_info per route:
model_info:
supports_reasoning: true
supports_none_reasoning_effort: true
supports_minimal_reasoning_effort: false
supports_xhigh_reasoning_effort: true
supports_max_reasoning_effort: true
which duplicates data LiteLLM already carries for the openai/ and azure/ spellings of the same model.
Secondary: supports_max_reasoning_effort missing on all gpt-5.6 variants
Separate from the chatgpt/ gap — no gpt-5.6 entry under any prefix declares supports_max_reasoning_effort, though the models do accept effort: "max". models.dev reports for openai/gpt-5.6-sol:
"reasoning_options": [
{ "type": "effort", "values": ["none", "low", "medium", "high", "xhigh", "max"] }
]
Verified end-to-end against a live proxy — reasoning_effort: "max" against a chatgpt/gpt-5.6-sol route is accepted and returns reasoning_tokens > 0.
For reference, the flag is well-established in the map (supports_max_reasoning_effort appears on 32 entries, supports_xhigh_reasoning_effort on 63), so this is a per-model omission rather than a schema question.
Suggested fix
Propagate the reasoning annotations from the openai/ entries to their chatgpt/ counterparts, and add supports_max_reasoning_effort: true to the gpt-5.6 family. For chatgpt/gpt-5.6-{sol,terra,luna}:
{
"supports_reasoning": true,
"supports_none_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"supports_xhigh_reasoning_effort": true,
"supports_max_reasoning_effort": true
}
(medium/high intentionally omitted — consistent with the rest of the map, which never declares them and treats them as the implicit baseline.)
Happy to open a PR for the data change if that's the preferred route.
Environment
- LiteLLM 1.86.0, self-hosted proxy, ChatGPT/Codex OAuth backend (
chatgptprovider,mode: responses) - Cost map fetched live from
main(4030 entries) rather than the bundled backup (2716 entries) — the gap is present in both
The two patches
litellm/model_prices_and_context_window_backup.jsonindex 56129fef13..5758677404 100644@@ -30551,7 +30551,11 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true,+ "supports_none_reasoning_effort": true,+ "supports_minimal_reasoning_effort": false,+ "supports_xhigh_reasoning_effort": true }, "chatgpt/gpt-5.6-luna": { "litellm_provider": "chatgpt",@@ -30567,7 +30571,12 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true,+ "supports_none_reasoning_effort": true,+ "supports_minimal_reasoning_effort": false,+ "supports_xhigh_reasoning_effort": true,+ "supports_max_reasoning_effort": true }, "chatgpt/gpt-5.6-sol": { "litellm_provider": "chatgpt",@@ -30583,7 +30592,12 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true,+ "supports_none_reasoning_effort": true,+ "supports_minimal_reasoning_effort": false,+ "supports_xhigh_reasoning_effort": true,+ "supports_max_reasoning_effort": true }, "chatgpt/gpt-5.6-terra": { "litellm_provider": "chatgpt",@@ -30599,7 +30613,12 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true,+ "supports_none_reasoning_effort": true,+ "supports_minimal_reasoning_effort": false,+ "supports_xhigh_reasoning_effort": true,+ "supports_max_reasoning_effort": true }, "chatgpt/gpt-5.4": { "litellm_provider": "chatgpt",@@ -30614,7 +30633,11 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true,+ "supports_none_reasoning_effort": true,+ "supports_minimal_reasoning_effort": false,+ "supports_xhigh_reasoning_effort": true }, "chatgpt/gpt-5.4-pro": { "litellm_provider": "chatgpt",@@ -30628,7 +30651,11 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true,+ "supports_none_reasoning_effort": false,+ "supports_minimal_reasoning_effort": false,+ "supports_xhigh_reasoning_effort": true }, "chatgpt/gpt-5.3-codex": { "litellm_provider": "chatgpt",@@ -30642,7 +30669,11 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true,+ "supports_none_reasoning_effort": false,+ "supports_minimal_reasoning_effort": true,+ "supports_xhigh_reasoning_effort": false }, "chatgpt/gpt-5.3-codex-spark": { "litellm_provider": "chatgpt",@@ -30656,7 +30687,8 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true }, "chatgpt/gpt-5.3-instant": { "litellm_provider": "chatgpt",@@ -30671,7 +30703,8 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true }, "chatgpt/gpt-5.3-chat-latest": { "litellm_provider": "chatgpt",@@ -30686,7 +30719,8 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true }, "chatgpt/gpt-5.2-codex": { "litellm_provider": "chatgpt",@@ -30700,7 +30734,8 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true }, "chatgpt/gpt-5.2": { "litellm_provider": "chatgpt",@@ -30715,7 +30750,11 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true,+ "supports_none_reasoning_effort": true,+ "supports_minimal_reasoning_effort": false,+ "supports_xhigh_reasoning_effort": true }, "chatgpt/gpt-5.1-codex-max": { "litellm_provider": "chatgpt",@@ -30729,7 +30768,8 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true }, "chatgpt/gpt-5.1-codex-mini": { "litellm_provider": "chatgpt",@@ -30743,7 +30783,8 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true }, "gigachat/GigaChat-2": { "input_cost_per_token": 0.0,litellm/router_utils/reasoning_effort_capability.pyindex 1d7656f253..2727640e3e 100644@@ -10,11 +10,12 @@ opt-in. none is opt-out everywhere except the azure gpt-5 family, whose config r UnsupportedParamsError without an explicit true. xhigh is gated on the request path by the openai and azure gpt-5 configs. max is not gated there at-all: outside the gpt-6-astra rows every entry carrying supports_max_reasoning_effort is Claude-family,-and anthropic/chat/transformation.py gates max on the output_config path while its reasoning_effort-path maps any level to a thinking budget. Making max opt-in is a deliberate trade, then, since an-explicit flag is the only signal that the tier is a real one rather than litellm rounding the level-to a budget, and a missing flag costs advisory metadata rather than a rejected request.+all: outside the OpenAI gpt-6 rows and the bedrock and chatgpt gpt-5.6 rows every entry carrying+supports_max_reasoning_effort is Claude-family, and anthropic/chat/transformation.py gates+max on the output_config path while its reasoning_effort path maps any level to a thinking budget.+Making max opt-in is a deliberate trade, then, since an explicit flag is the only signal that the+tier is a real one rather than litellm rounding the level to a budget, and a missing flag costs+advisory metadata rather than a rejected request. A deployment the map describes with no effort flags at all resolves to None rather than to the opt-out defaults. 689 of the map's 854 reasoning entries carry no flag, and the o-series, xai andmodel_prices_and_context_window.jsonindex 56129fef13..5758677404 100644@@ -30551,7 +30551,11 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true,+ "supports_none_reasoning_effort": true,+ "supports_minimal_reasoning_effort": false,+ "supports_xhigh_reasoning_effort": true }, "chatgpt/gpt-5.6-luna": { "litellm_provider": "chatgpt",@@ -30567,7 +30571,12 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true,+ "supports_none_reasoning_effort": true,+ "supports_minimal_reasoning_effort": false,+ "supports_xhigh_reasoning_effort": true,+ "supports_max_reasoning_effort": true }, "chatgpt/gpt-5.6-sol": { "litellm_provider": "chatgpt",@@ -30583,7 +30592,12 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true,+ "supports_none_reasoning_effort": true,+ "supports_minimal_reasoning_effort": false,+ "supports_xhigh_reasoning_effort": true,+ "supports_max_reasoning_effort": true }, "chatgpt/gpt-5.6-terra": { "litellm_provider": "chatgpt",@@ -30599,7 +30613,12 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true,+ "supports_none_reasoning_effort": true,+ "supports_minimal_reasoning_effort": false,+ "supports_xhigh_reasoning_effort": true,+ "supports_max_reasoning_effort": true }, "chatgpt/gpt-5.4": { "litellm_provider": "chatgpt",@@ -30614,7 +30633,11 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true,+ "supports_none_reasoning_effort": true,+ "supports_minimal_reasoning_effort": false,+ "supports_xhigh_reasoning_effort": true }, "chatgpt/gpt-5.4-pro": { "litellm_provider": "chatgpt",@@ -30628,7 +30651,11 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true,+ "supports_none_reasoning_effort": false,+ "supports_minimal_reasoning_effort": false,+ "supports_xhigh_reasoning_effort": true }, "chatgpt/gpt-5.3-codex": { "litellm_provider": "chatgpt",@@ -30642,7 +30669,11 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true,+ "supports_none_reasoning_effort": false,+ "supports_minimal_reasoning_effort": true,+ "supports_xhigh_reasoning_effort": false }, "chatgpt/gpt-5.3-codex-spark": { "litellm_provider": "chatgpt",@@ -30656,7 +30687,8 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true }, "chatgpt/gpt-5.3-instant": { "litellm_provider": "chatgpt",@@ -30671,7 +30703,8 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true }, "chatgpt/gpt-5.3-chat-latest": { "litellm_provider": "chatgpt",@@ -30686,7 +30719,8 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true }, "chatgpt/gpt-5.2-codex": { "litellm_provider": "chatgpt",@@ -30700,7 +30734,8 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true }, "chatgpt/gpt-5.2": { "litellm_provider": "chatgpt",@@ -30715,7 +30750,11 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true,+ "supports_none_reasoning_effort": true,+ "supports_minimal_reasoning_effort": false,+ "supports_xhigh_reasoning_effort": true }, "chatgpt/gpt-5.1-codex-max": { "litellm_provider": "chatgpt",@@ -30729,7 +30768,8 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true }, "chatgpt/gpt-5.1-codex-mini": { "litellm_provider": "chatgpt",@@ -30743,7 +30783,8 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true }, "gigachat/GigaChat-2": { "input_cost_per_token": 0.0,tests/unit/test_model_prices_schema.pyindex 052278631e..1928c0c74f 100644@@ -328,6 +328,61 @@ def test_every_bedrock_openai_gpt_row_advertises_xhigh(prices: dict): assert missing == [] +def is_reasoning_capability_key(key: str) -> bool:+ return key == "supports_reasoning" or (key.startswith("supports_") and key.endswith("_reasoning_effort"))+++@pytest.mark.parametrize("path", (PRICES_PATH, BACKUP_PRICES_PATH), ids=("main", "backup"))+def test_chatgpt_rows_carry_the_reasoning_flags_of_their_openai_twin(path: Path):+ """chatgpt/ serves the same OpenAI models over the Codex backend, so /model/info for a+ chatgpt/ route must not report supports_reasoning or an effort flag as unknown when the openai+ spelling of the same model answers it. Every row is a gpt-5 reasoning model, so rows without an+ openai twin (codex-spark, instant) still carry supports_reasoning."""+ rows: Mapping[str, object] = json.loads(path.read_text())+ chatgpt_rows: Final = {+ name: entry for name, entry in rows.items() if isinstance(entry, dict) and name.startswith("chatgpt/")+ }+ drifted: Final = [+ f"{name}.{key}: {entry.get(key)!r} != openai {value!r}"+ for name, entry in chatgpt_rows.items()+ if isinstance(twin := rows.get(name.removeprefix("chatgpt/")), dict)+ for key, value in twin.items()+ if is_reasoning_capability_key(key) and entry.get(key) != value+ ]+ unflagged: Final = [name for name, entry in chatgpt_rows.items() if entry.get("supports_reasoning") is not True]+ assert chatgpt_rows+ assert drifted == []+ assert unflagged == []+++@pytest.mark.parametrize("variant", ("luna", "sol", "terra"))+def test_chatgpt_gpt_5_6_rows_advertise_max(prices: dict, variant: str):+ """The chatgpt provider only speaks the Codex Responses API, which accepts max for gpt-5.6+ (verified live through the proxy with reasoning_tokens > 0), unlike the direct+ /v1/chat/completions surface the bare gpt-5.6 rows describe."""+ name = f"chatgpt/gpt-5.6-{variant}"+ entry = dict(prices[name], key=name)+ assert resolve_supported_reasoning_efforts(entry, deployment_is_mapped=True) == (+ "none",+ "low",+ "medium",+ "high",+ "xhigh",+ "max",+ )+++def test_chatgpt_gpt_5_6_model_info_reports_reasoning_support():+ """get_model_info is what /model/info serves for a chatgpt/gpt-5.6-sol route, and clients gate+ their effort picker on these fields, so None here disabled reasoning controls entirely."""+ info = litellm.get_model_info("chatgpt/gpt-5.6-sol")+ assert info["supports_reasoning"] is True+ assert info["supports_none_reasoning_effort"] is True+ assert info["supports_minimal_reasoning_effort"] is False+ assert info["supports_xhigh_reasoning_effort"] is True+ assert info["supports_max_reasoning_effort"] is True++ def is_active_priced_mistral_chat_row(name: str, entry: Mapping[str, object]) -> bool: input_cost: Final = entry.get("input_cost_per_token") return (
litellm-rust/crates/model-catalog/src/model_info.rsindex 361cb56e9b1..9e9a4220e50 100644@@ -42,6 +42,9 @@ pub struct ModelInfo { pub cache_creation_input_token_cost_above_200k_tokens: Option<f64>, /// Rate applied once the prompt exceeds the token threshold in the field name. #[serde(skip_serializing_if = "Option::is_none")]+ pub cache_creation_input_token_cost_above_200k_tokens_batches: Option<f64>,+ /// Rate applied once the prompt exceeds the token threshold in the field name.+ #[serde(skip_serializing_if = "Option::is_none")] pub cache_creation_input_token_cost_above_256k_tokens: Option<f64>, /// Rate applied once the prompt exceeds the token threshold in the field name. #[serde(skip_serializing_if = "Option::is_none")]@@ -78,6 +81,9 @@ pub struct ModelInfo { /// Rate applied once the prompt exceeds the token threshold in the field name. #[serde(skip_serializing_if = "Option::is_none")] pub cache_read_input_token_cost_above_200k_tokens: Option<f64>,+ /// Rate applied once the prompt exceeds the token threshold in the field name.+ #[serde(skip_serializing_if = "Option::is_none")]+ pub cache_read_input_token_cost_above_200k_tokens_batches: Option<f64>, /// Priority service-tier rate for the same-named base field. #[serde(skip_serializing_if = "Option::is_none")] pub cache_read_input_token_cost_above_200k_tokens_priority: Option<f64>,@@ -113,6 +119,10 @@ pub struct ModelInfo { pub code_interpreter_cost_per_session: Option<f64>, #[serde(skip_serializing_if = "Option::is_none")] pub comment: Option<String>,+ #[serde(skip_serializing_if = "Option::is_none")]+ pub computer_use_input_cost_per_1k_tokens: Option<f64>,+ #[serde(skip_serializing_if = "Option::is_none")]+ pub computer_use_output_cost_per_1k_tokens: Option<f64>, /// Reasoning effort the provider applies when the request omits reasoning_effort. Gates whether a non-default temperature or the top_p/logprobs sampling params are accepted, which hold only when the effort resolves to 'none'. #[serde(skip_serializing_if = "Option::is_none")] pub default_reasoning_effort: Option<ReasoningEffort>,@@ -120,6 +130,10 @@ pub struct ModelInfo { #[serde(skip_serializing_if = "Option::is_none")] pub deprecation_date: Option<String>, #[serde(skip_serializing_if = "Option::is_none")]+ pub file_search_cost_per_1k_calls: Option<f64>,+ #[serde(skip_serializing_if = "Option::is_none")]+ pub file_search_cost_per_gb_per_day: Option<f64>,+ #[serde(skip_serializing_if = "Option::is_none")] pub gemini_audio_only_live: Option<bool>, #[serde(skip_serializing_if = "Option::is_none")] pub gemini_native_audio: Option<bool>,@@ -174,6 +188,9 @@ pub struct ModelInfo { /// Rate applied once the prompt exceeds the token threshold in the field name. #[serde(skip_serializing_if = "Option::is_none")] pub input_cost_per_token_above_200k_tokens: Option<f64>,+ /// Rate applied once the prompt exceeds the token threshold in the field name.+ #[serde(skip_serializing_if = "Option::is_none")]+ pub input_cost_per_token_above_200k_tokens_batches: Option<f64>, /// Priority service-tier rate for the same-named base field. #[serde(skip_serializing_if = "Option::is_none")] pub input_cost_per_token_above_200k_tokens_priority: Option<f64>,@@ -265,6 +282,26 @@ pub struct ModelInfo { pub output_cost_per_image_1536: Option<f64>, #[serde(skip_serializing_if = "Option::is_none")] pub output_cost_per_image_512: Option<f64>,+ #[serde(+ rename = "output_cost_per_image_0.5K",+ skip_serializing_if = "Option::is_none"+ )]+ pub output_cost_per_image_0_5k: Option<f64>,+ #[serde(+ rename = "output_cost_per_image_1K",+ skip_serializing_if = "Option::is_none"+ )]+ pub output_cost_per_image_1k: Option<f64>,+ #[serde(+ rename = "output_cost_per_image_2K",+ skip_serializing_if = "Option::is_none"+ )]+ pub output_cost_per_image_2k: Option<f64>,+ #[serde(+ rename = "output_cost_per_image_4K",+ skip_serializing_if = "Option::is_none"+ )]+ pub output_cost_per_image_4k: Option<f64>, #[serde(skip_serializing_if = "Option::is_none")] pub output_cost_per_image_token: Option<f64>, #[serde(skip_serializing_if = "Option::is_none")]@@ -297,6 +334,9 @@ pub struct ModelInfo { /// Rate applied once the prompt exceeds the token threshold in the field name. #[serde(skip_serializing_if = "Option::is_none")] pub output_cost_per_token_above_200k_tokens: Option<f64>,+ /// Rate applied once the prompt exceeds the token threshold in the field name.+ #[serde(skip_serializing_if = "Option::is_none")]+ pub output_cost_per_token_above_200k_tokens_batches: Option<f64>, /// Priority service-tier rate for the same-named base field. #[serde(skip_serializing_if = "Option::is_none")] pub output_cost_per_token_above_200k_tokens_priority: Option<f64>,@@ -357,6 +397,8 @@ pub struct ModelInfo { /// Provider default requests-per-minute limit. #[serde(skip_serializing_if = "Option::is_none")] pub rpm: Option<u64>,+ #[serde(skip_serializing_if = "Option::is_none")]+ pub rules: Option<Vec<Value>>, /// USD cost per web search query, keyed by search context size. #[serde(skip_serializing_if = "Option::is_none")] pub search_context_cost_per_query: Option<SearchContextCostPerQuery>,@@ -475,6 +517,8 @@ pub struct ModelInfo { #[serde(skip_serializing_if = "Option::is_none")] pub uses_embed_content: Option<bool>, #[serde(skip_serializing_if = "Option::is_none")]+ pub vector_store_cost_per_gb_per_day: Option<f64>,+ #[serde(skip_serializing_if = "Option::is_none")] pub vertex_ai_audio_api: Option<VertexAiAudioApi>, /// Whether web search is billed per query or per prompt. #[serde(skip_serializing_if = "Option::is_none")]litellm/litellm_core_utils/litellm_logging.pyindex 28d72702f3e..e955c0157c6 100644@@ -384,6 +384,7 @@ _DEPLOYMENT_PRICING_KEYS: Final = ( "cache_read_input_token_cost_above_200k_tokens_batches", "cache_read_input_token_cost_above_272k_tokens_batches", "cache_creation_input_token_cost_batches",+ "cache_creation_input_token_cost_above_200k_tokens_batches", "cache_creation_input_token_cost_above_272k_tokens_batches", "ocr_cost_per_page", "ocr_cost_per_page_batches",litellm/model_prices_and_context_window_backup.jsonindex 56129fef135..ec31039fee0 100644@@ -1279,8 +1279,11 @@ "source": "https://pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrockFoundationModels/current/index.json" }, "anthropic.claude-mythos-preview": {- "input_cost_per_token": 0,- "output_cost_per_token": 0,+ "cache_creation_input_token_cost": 3.4375e-05,+ "cache_creation_input_token_cost_above_1hr": 5.5e-05,+ "cache_read_input_token_cost": 2.75e-06,+ "input_cost_per_token": 2.75e-05,+ "output_cost_per_token": 0.0001375, "litellm_provider": "bedrock", "max_input_tokens": 1000000, "max_output_tokens": 128000,@@ -1289,10 +1292,11 @@ "thinking_always_on": true, "supports_function_calling": true, "supports_vision": true,- "supports_prompt_caching": false,+ "supports_prompt_caching": true, "supports_reasoning": true, "supports_tool_choice": true,- "supports_output_config": true+ "supports_output_config": true,+ "source": "https://pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrockFoundationModels/current/index.json" }, "global.anthropic.claude-opus-4-7": { "bedrock_converse_supports_strict_tools": false,@@ -7773,27 +7777,27 @@ "output_cost_per_token_above_272k_tokens_batches": 0.000135 }, "azure/gpt-5.6": {- "cache_creation_input_token_cost": 6.25e-06,- "cache_creation_input_token_cost_above_272k_tokens": 1.25e-05,- "cache_creation_input_token_cost_priority": 1.25e-05,- "cache_creation_input_token_cost_above_272k_tokens_priority": 2.5e-05,- "cache_read_input_token_cost": 5e-07,- "cache_read_input_token_cost_above_272k_tokens": 1e-06,- "cache_read_input_token_cost_priority": 1e-06,- "cache_read_input_token_cost_above_272k_tokens_priority": 2e-06,- "input_cost_per_token": 5e-06,- "input_cost_per_token_above_272k_tokens": 1e-05,- "input_cost_per_token_priority": 1e-05,- "input_cost_per_token_above_272k_tokens_priority": 2e-05,+ "cache_creation_input_token_cost": 5e-06,+ "cache_creation_input_token_cost_above_272k_tokens": 1e-05,+ "cache_creation_input_token_cost_priority": 1e-05,+ "cache_creation_input_token_cost_above_272k_tokens_priority": 2e-05,+ "cache_read_input_token_cost": 4e-07,+ "cache_read_input_token_cost_above_272k_tokens": 8e-07,+ "cache_read_input_token_cost_priority": 8e-07,+ "cache_read_input_token_cost_above_272k_tokens_priority": 1.6e-06,+ "input_cost_per_token": 4e-06,+ "input_cost_per_token_above_272k_tokens": 8e-06,+ "input_cost_per_token_priority": 8e-06,+ "input_cost_per_token_above_272k_tokens_priority": 1.6e-05, "litellm_provider": "azure", "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat",- "output_cost_per_token": 3e-05,- "output_cost_per_token_above_272k_tokens": 4.5e-05,- "output_cost_per_token_priority": 6e-05,- "output_cost_per_token_above_272k_tokens_priority": 9e-05,+ "output_cost_per_token": 2e-05,+ "output_cost_per_token_above_272k_tokens": 3e-05,+ "output_cost_per_token_priority": 4e-05,+ "output_cost_per_token_above_272k_tokens_priority": 6e-05, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -8579,27 +8583,27 @@ "supports_web_search": true }, "azure/us/gpt-5.6": {- "cache_creation_input_token_cost": 6.875e-06,- "cache_creation_input_token_cost_above_272k_tokens": 1.375e-05,- "cache_creation_input_token_cost_above_272k_tokens_priority": 2.75e-05,- "cache_creation_input_token_cost_priority": 1.375e-05,- "cache_read_input_token_cost": 5.5e-07,- "cache_read_input_token_cost_above_272k_tokens": 1.1e-06,- "cache_read_input_token_cost_above_272k_tokens_priority": 2.2e-06,- "cache_read_input_token_cost_priority": 1.1e-06,- "input_cost_per_token": 5.5e-06,- "input_cost_per_token_above_272k_tokens": 1.1e-05,- "input_cost_per_token_above_272k_tokens_priority": 2.2e-05,- "input_cost_per_token_priority": 1.1e-05,+ "cache_creation_input_token_cost": 5.5e-06,+ "cache_creation_input_token_cost_above_272k_tokens": 1.1e-05,+ "cache_creation_input_token_cost_above_272k_tokens_priority": 2.2e-05,+ "cache_creation_input_token_cost_priority": 1.1e-05,+ "cache_read_input_token_cost": 4.4e-07,+ "cache_read_input_token_cost_above_272k_tokens": 8.8e-07,+ "cache_read_input_token_cost_above_272k_tokens_priority": 1.76e-06,+ "cache_read_input_token_cost_priority": 8.8e-07,+ "input_cost_per_token": 4.4e-06,+ "input_cost_per_token_above_272k_tokens": 8.8e-06,+ "input_cost_per_token_above_272k_tokens_priority": 1.76e-05,+ "input_cost_per_token_priority": 8.8e-06, "litellm_provider": "azure", "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat",- "output_cost_per_token": 3.3e-05,- "output_cost_per_token_above_272k_tokens": 4.95e-05,- "output_cost_per_token_above_272k_tokens_priority": 9.9e-05,- "output_cost_per_token_priority": 6.6e-05,+ "output_cost_per_token": 2.2e-05,+ "output_cost_per_token_above_272k_tokens": 3.3e-05,+ "output_cost_per_token_above_272k_tokens_priority": 6.6e-05,+ "output_cost_per_token_priority": 4.4e-05, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -8985,27 +8989,27 @@ "supports_web_search": true }, "azure/eu/gpt-5.6": {- "cache_creation_input_token_cost": 6.875e-06,- "cache_creation_input_token_cost_above_272k_tokens": 1.375e-05,- "cache_creation_input_token_cost_above_272k_tokens_priority": 2.75e-05,- "cache_creation_input_token_cost_priority": 1.375e-05,- "cache_read_input_token_cost": 5.5e-07,- "cache_read_input_token_cost_above_272k_tokens": 1.1e-06,- "cache_read_input_token_cost_above_272k_tokens_priority": 2.2e-06,- "cache_read_input_token_cost_priority": 1.1e-06,- "input_cost_per_token": 5.5e-06,- "input_cost_per_token_above_272k_tokens": 1.1e-05,- "input_cost_per_token_above_272k_tokens_priority": 2.2e-05,- "input_cost_per_token_priority": 1.1e-05,+ "cache_creation_input_token_cost": 5.5e-06,+ "cache_creation_input_token_cost_above_272k_tokens": 1.1e-05,+ "cache_creation_input_token_cost_above_272k_tokens_priority": 2.2e-05,+ "cache_creation_input_token_cost_priority": 1.1e-05,+ "cache_read_input_token_cost": 4.4e-07,+ "cache_read_input_token_cost_above_272k_tokens": 8.8e-07,+ "cache_read_input_token_cost_above_272k_tokens_priority": 1.76e-06,+ "cache_read_input_token_cost_priority": 8.8e-07,+ "input_cost_per_token": 4.4e-06,+ "input_cost_per_token_above_272k_tokens": 8.8e-06,+ "input_cost_per_token_above_272k_tokens_priority": 1.76e-05,+ "input_cost_per_token_priority": 8.8e-06, "litellm_provider": "azure", "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat",- "output_cost_per_token": 3.3e-05,- "output_cost_per_token_above_272k_tokens": 4.95e-05,- "output_cost_per_token_above_272k_tokens_priority": 9.9e-05,- "output_cost_per_token_priority": 6.6e-05,+ "output_cost_per_token": 2.2e-05,+ "output_cost_per_token_above_272k_tokens": 3.3e-05,+ "output_cost_per_token_above_272k_tokens_priority": 6.6e-05,+ "output_cost_per_token_priority": 4.4e-05, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -14640,14 +14644,18 @@ "claude-haiku-4-5-20251001": { "cache_creation_input_token_cost": 1.25e-06, "cache_creation_input_token_cost_above_1hr": 2e-06,+ "cache_creation_input_token_cost_batches": 6.25e-07, "cache_read_input_token_cost": 1e-07,+ "cache_read_input_token_cost_batches": 5e-08, "input_cost_per_token": 1e-06,+ "input_cost_per_token_batches": 5e-07, "litellm_provider": "anthropic", "max_input_tokens": 200000, "max_output_tokens": 64000, "max_tokens": 64000, "mode": "chat", "output_cost_per_token": 5e-06,+ "output_cost_per_token_batches": 2.5e-06, "supports_assistant_prefill": true, "supports_function_calling": true, "supports_native_structured_output": true,@@ -14663,14 +14671,18 @@ "claude-haiku-4-5": { "cache_creation_input_token_cost": 1.25e-06, "cache_creation_input_token_cost_above_1hr": 2e-06,+ "cache_creation_input_token_cost_batches": 6.25e-07, "cache_read_input_token_cost": 1e-07,+ "cache_read_input_token_cost_batches": 5e-08, "input_cost_per_token": 1e-06,+ "input_cost_per_token_batches": 5e-07, "litellm_provider": "anthropic", "max_input_tokens": 200000, "max_output_tokens": 64000, "max_tokens": 64000, "mode": "chat", "output_cost_per_token": 5e-06,+ "output_cost_per_token_batches": 2.5e-06, "supports_assistant_prefill": true, "supports_function_calling": true, "supports_native_structured_output": true,@@ -14693,13 +14705,21 @@ "input_cost_per_token_above_200k_tokens": 6e-06, "output_cost_per_token_above_200k_tokens": 2.25e-05, "cache_creation_input_token_cost_above_200k_tokens": 7.5e-06,+ "cache_creation_input_token_cost_above_200k_tokens_batches": 3.75e-06,+ "cache_creation_input_token_cost_batches": 1.875e-06, "cache_read_input_token_cost_above_200k_tokens": 6e-07,+ "cache_read_input_token_cost_above_200k_tokens_batches": 3e-07,+ "cache_read_input_token_cost_batches": 1.5e-07,+ "input_cost_per_token_above_200k_tokens_batches": 3e-06,+ "input_cost_per_token_batches": 1.5e-06, "litellm_provider": "anthropic", "max_input_tokens": 1000000, "max_output_tokens": 64000, "max_tokens": 64000, "mode": "chat", "output_cost_per_token": 1.5e-05,+ "output_cost_per_token_above_200k_tokens_batches": 1.125e-05,+ "output_cost_per_token_batches": 7.5e-06, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -14727,13 +14747,21 @@ "input_cost_per_token_above_200k_tokens": 6e-06, "output_cost_per_token_above_200k_tokens": 2.25e-05, "cache_creation_input_token_cost_above_200k_tokens": 7.5e-06,+ "cache_creation_input_token_cost_above_200k_tokens_batches": 3.75e-06,+ "cache_creation_input_token_cost_batches": 1.875e-06, "cache_read_input_token_cost_above_200k_tokens": 6e-07,+ "cache_read_input_token_cost_above_200k_tokens_batches": 3e-07,+ "cache_read_input_token_cost_batches": 1.5e-07,+ "input_cost_per_token_above_200k_tokens_batches": 3e-06,+ "input_cost_per_token_batches": 1.5e-06, "litellm_provider": "anthropic", "max_input_tokens": 1000000, "max_output_tokens": 64000, "max_tokens": 64000, "mode": "chat", "output_cost_per_token": 1.5e-05,+ "output_cost_per_token_above_200k_tokens_batches": 1.125e-05,+ "output_cost_per_token_batches": 7.5e-06, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -14757,14 +14785,18 @@ "supports_anthropic_compaction": true, "cache_creation_input_token_cost": 2.5e-06, "cache_creation_input_token_cost_above_1hr": 4e-06,+ "cache_creation_input_token_cost_batches": 1.25e-06, "cache_read_input_token_cost": 2e-07,+ "cache_read_input_token_cost_batches": 1e-07, "input_cost_per_token": 2e-06,+ "input_cost_per_token_batches": 1e-06, "litellm_provider": "anthropic", "max_input_tokens": 1000000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 1e-05,+ "output_cost_per_token_batches": 5e-06, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -14797,14 +14829,18 @@ "supports_anthropic_compaction": true, "cache_creation_input_token_cost": 3.75e-06, "cache_creation_input_token_cost_above_1hr": 6e-06,+ "cache_creation_input_token_cost_batches": 1.875e-06, "cache_read_input_token_cost": 3e-07,+ "cache_read_input_token_cost_batches": 1.5e-07, "input_cost_per_token": 3e-06,+ "input_cost_per_token_batches": 1.5e-06, "litellm_provider": "anthropic", "max_input_tokens": 1000000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 1.5e-05,+ "output_cost_per_token_batches": 7.5e-06, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -14865,14 +14901,18 @@ "claude-opus-4-5-20251101": { "cache_creation_input_token_cost": 6.25e-06, "cache_creation_input_token_cost_above_1hr": 1e-05,+ "cache_creation_input_token_cost_batches": 3.125e-06, "cache_read_input_token_cost": 5e-07,+ "cache_read_input_token_cost_batches": 2.5e-07, "input_cost_per_token": 5e-06,+ "input_cost_per_token_batches": 2.5e-06, "litellm_provider": "anthropic", "max_input_tokens": 200000, "max_output_tokens": 64000, "max_tokens": 64000, "mode": "chat", "output_cost_per_token": 2.5e-05,+ "output_cost_per_token_batches": 1.25e-05, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -14895,14 +14935,18 @@ "claude-opus-4-5": { "cache_creation_input_token_cost": 6.25e-06, "cache_creation_input_token_cost_above_1hr": 1e-05,+ "cache_creation_input_token_cost_batches": 3.125e-06, "cache_read_input_token_cost": 5e-07,+ "cache_read_input_token_cost_batches": 2.5e-07, "input_cost_per_token": 5e-06,+ "input_cost_per_token_batches": 2.5e-06, "litellm_provider": "anthropic", "max_input_tokens": 200000, "max_output_tokens": 64000, "max_tokens": 64000, "mode": "chat", "output_cost_per_token": 2.5e-05,+ "output_cost_per_token_batches": 1.25e-05, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -14927,14 +14971,18 @@ "supports_anthropic_compaction": true, "cache_creation_input_token_cost": 6.25e-06, "cache_creation_input_token_cost_above_1hr": 1e-05,+ "cache_creation_input_token_cost_batches": 3.125e-06, "cache_read_input_token_cost": 5e-07,+ "cache_read_input_token_cost_batches": 2.5e-07, "input_cost_per_token": 5e-06,+ "input_cost_per_token_batches": 2.5e-06, "litellm_provider": "anthropic", "max_input_tokens": 1000000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 2.5e-05,+ "output_cost_per_token_batches": 1.25e-05, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -14966,14 +15014,18 @@ "supports_anthropic_compaction": true, "cache_creation_input_token_cost": 6.25e-06, "cache_creation_input_token_cost_above_1hr": 1e-05,+ "cache_creation_input_token_cost_batches": 3.125e-06, "cache_read_input_token_cost": 5e-07,+ "cache_read_input_token_cost_batches": 2.5e-07, "input_cost_per_token": 5e-06,+ "input_cost_per_token_batches": 2.5e-06, "litellm_provider": "anthropic", "max_input_tokens": 1000000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 2.5e-05,+ "output_cost_per_token_batches": 1.25e-05, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -15004,14 +15056,18 @@ "supports_anthropic_compaction": true, "cache_creation_input_token_cost": 6.25e-06, "cache_creation_input_token_cost_above_1hr": 1e-05,+ "cache_creation_input_token_cost_batches": 3.125e-06, "cache_read_input_token_cost": 5e-07,+ "cache_read_input_token_cost_batches": 2.5e-07, "input_cost_per_token": 5e-06,+ "input_cost_per_token_batches": 2.5e-06, "litellm_provider": "anthropic", "max_input_tokens": 1000000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 2.5e-05,+ "output_cost_per_token_batches": 1.25e-05, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -15044,14 +15100,18 @@ "supports_anthropic_compaction": true, "cache_creation_input_token_cost": 6.25e-06, "cache_creation_input_token_cost_above_1hr": 1e-05,+ "cache_creation_input_token_cost_batches": 3.125e-06, "cache_read_input_token_cost": 5e-07,+ "cache_read_input_token_cost_batches": 2.5e-07, "input_cost_per_token": 5e-06,+ "input_cost_per_token_batches": 2.5e-06, "litellm_provider": "anthropic", "max_input_tokens": 1000000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 2.5e-05,+ "output_cost_per_token_batches": 1.25e-05, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -15083,14 +15143,18 @@ "supports_anthropic_compaction": true, "cache_creation_input_token_cost": 1.25e-05, "cache_creation_input_token_cost_above_1hr": 2e-05,+ "cache_creation_input_token_cost_batches": 6.25e-06, "cache_read_input_token_cost": 1e-06,+ "cache_read_input_token_cost_batches": 5e-07, "input_cost_per_token": 1e-05,+ "input_cost_per_token_batches": 5e-06, "litellm_provider": "anthropic", "max_input_tokens": 1000000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 5e-05,+ "output_cost_per_token_batches": 2.5e-05, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -15123,14 +15187,18 @@ "supports_anthropic_compaction": true, "cache_creation_input_token_cost": 1.25e-05, "cache_creation_input_token_cost_above_1hr": 2e-05,+ "cache_creation_input_token_cost_batches": 6.25e-06, "cache_read_input_token_cost": 2.5e-07,+ "cache_read_input_token_cost_batches": 1.25e-07, "input_cost_per_token": 1e-05,+ "input_cost_per_token_batches": 5e-06, "litellm_provider": "anthropic", "max_input_tokens": 1000000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 5e-05,+ "output_cost_per_token_batches": 2.5e-05, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -15164,14 +15232,18 @@ "supports_anthropic_compaction": true, "cache_creation_input_token_cost": 5e-06, "cache_creation_input_token_cost_above_1hr": 8e-06,+ "cache_creation_input_token_cost_batches": 2.5e-06, "cache_read_input_token_cost": 2e-07,+ "cache_read_input_token_cost_batches": 1e-07, "input_cost_per_token": 4e-06,+ "input_cost_per_token_batches": 2e-06, "litellm_provider": "anthropic", "max_input_tokens": 1000000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 2e-05,+ "output_cost_per_token_batches": 1e-05, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -15207,14 +15279,18 @@ "supports_anthropic_compaction": true, "cache_creation_input_token_cost": 6.25e-06, "cache_creation_input_token_cost_above_1hr": 1e-05,+ "cache_creation_input_token_cost_batches": 3.125e-06, "cache_read_input_token_cost": 5e-07,+ "cache_read_input_token_cost_batches": 2.5e-07, "input_cost_per_token": 5e-06,+ "input_cost_per_token_batches": 2.5e-06, "litellm_provider": "anthropic", "max_input_tokens": 1000000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 2.5e-05,+ "output_cost_per_token_batches": 1.25e-05, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -15250,14 +15326,18 @@ "supports_anthropic_compaction": true, "cache_creation_input_token_cost": 6.25e-06, "cache_creation_input_token_cost_above_1hr": 1e-05,+ "cache_creation_input_token_cost_batches": 3.125e-06, "cache_read_input_token_cost": 5e-07,+ "cache_read_input_token_cost_batches": 2.5e-07, "input_cost_per_token": 5e-06,+ "input_cost_per_token_batches": 2.5e-06, "litellm_provider": "anthropic", "max_input_tokens": 1000000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 2.5e-05,+ "output_cost_per_token_batches": 1.25e-05, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -15756,6 +15836,16 @@ "supports_function_calling": true, "supports_tool_choice": true },+ "c4ai-aya-expanse-32b": {+ "input_cost_per_token": 5e-07,+ "output_cost_per_token": 1.5e-06,+ "litellm_provider": "cohere_chat",+ "max_input_tokens": 128000,+ "max_output_tokens": 4000,+ "max_tokens": 4000,+ "mode": "chat",+ "source": "https://docs.cohere.com/docs/models"+ }, "command-a-plus-05-2026": { "input_cost_per_token": 0.0, "litellm_provider": "cohere_chat",@@ -22145,11 +22235,11 @@ } }, "bing_grounding/search": {- "input_cost_per_query": 0.035,+ "input_cost_per_query": 0.014, "litellm_provider": "bing_grounding", "mode": "search", "metadata": {- "notes": "Grounding with Bing Search (G1 SKU): $35 per 1,000 transactions. Tokens for the Foundry model deployment that runs the grounded search are billed separately on that deployment."+ "notes": "Grounding with Bing Search (G1 SKU): $14 per 1,000 transactions. Tokens for the Foundry model deployment that runs the grounded search are billed separately on that deployment." } }, "tinyfish/search": {@@ -28036,6 +28126,7 @@ "tpm": 10000000 }, "gemini/gemini-3-pro-image-preview": {+ "deprecation_date": "2026-06-25", "input_cost_per_image": 0.0011, "input_cost_per_token": 2e-06, "input_cost_per_token_batches": 1e-06,@@ -28083,6 +28174,7 @@ "supports_reasoning": false }, "gemini/gemini-3.1-flash-image-preview": {+ "deprecation_date": "2026-06-25", "input_cost_per_token": 5e-07, "input_cost_per_token_batches": 2.5e-07, "litellm_provider": "gemini",@@ -28130,6 +28222,7 @@ "cache_read_input_token_cost_batches": 1.25e-08, "cache_read_input_token_cost_flex": 1.25e-08, "cache_read_input_token_cost_priority": 4.5e-08,+ "deprecation_date": "2026-05-25", "input_cost_per_audio_token": 5e-07, "input_cost_per_token": 2.5e-07, "input_cost_per_token_batches": 1.25e-07,@@ -30551,7 +30644,11 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true,+ "supports_none_reasoning_effort": true,+ "supports_xhigh_reasoning_effort": true,+ "supports_minimal_reasoning_effort": false }, "chatgpt/gpt-5.6-luna": { "litellm_provider": "chatgpt",@@ -30567,7 +30664,11 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_minimal_reasoning_effort": false,+ "supports_none_reasoning_effort": true,+ "supports_reasoning": true,+ "supports_xhigh_reasoning_effort": true }, "chatgpt/gpt-5.6-sol": { "litellm_provider": "chatgpt",@@ -30583,7 +30684,11 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_minimal_reasoning_effort": false,+ "supports_none_reasoning_effort": true,+ "supports_reasoning": true,+ "supports_xhigh_reasoning_effort": true }, "chatgpt/gpt-5.6-terra": { "litellm_provider": "chatgpt",@@ -30599,7 +30704,11 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_minimal_reasoning_effort": false,+ "supports_none_reasoning_effort": true,+ "supports_reasoning": true,+ "supports_xhigh_reasoning_effort": true }, "chatgpt/gpt-5.4": { "litellm_provider": "chatgpt",@@ -30614,7 +30723,12 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true,+ "supports_none_reasoning_effort": true,+ "default_reasoning_effort": "none",+ "supports_xhigh_reasoning_effort": true,+ "supports_minimal_reasoning_effort": false }, "chatgpt/gpt-5.4-pro": { "litellm_provider": "chatgpt",@@ -30628,7 +30742,11 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true,+ "supports_none_reasoning_effort": false,+ "supports_xhigh_reasoning_effort": true,+ "supports_minimal_reasoning_effort": false }, "chatgpt/gpt-5.3-codex": { "litellm_provider": "chatgpt",@@ -30642,7 +30760,11 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true,+ "supports_none_reasoning_effort": false,+ "supports_xhigh_reasoning_effort": false,+ "supports_minimal_reasoning_effort": true }, "chatgpt/gpt-5.3-codex-spark": { "litellm_provider": "chatgpt",@@ -30686,7 +30808,8 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true }, "chatgpt/gpt-5.2-codex": { "litellm_provider": "chatgpt",@@ -30700,7 +30823,8 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true }, "chatgpt/gpt-5.2": { "litellm_provider": "chatgpt",@@ -30715,7 +30839,12 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true,+ "supports_none_reasoning_effort": true,+ "default_reasoning_effort": "none",+ "supports_xhigh_reasoning_effort": true,+ "supports_minimal_reasoning_effort": false }, "chatgpt/gpt-5.1-codex-max": { "litellm_provider": "chatgpt",@@ -30729,7 +30858,8 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true }, "chatgpt/gpt-5.1-codex-mini": { "litellm_provider": "chatgpt",@@ -30743,7 +30873,8 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true }, "gigachat/GigaChat-2": { "input_cost_per_token": 0.0,@@ -39214,6 +39345,17 @@ "supports_function_calling": true, "supports_reasoning": true },+ "nebius/deepseek-ai/DeepSeek-V4.1-Flash": {+ "input_cost_per_token": 3e-07,+ "litellm_provider": "nebius",+ "max_input_tokens": 1048576,+ "max_output_tokens": 1048576,+ "max_tokens": 1048576,+ "mode": "chat",+ "output_cost_per_token": 1.2e-06,+ "source": "https://tokenfactory.nebius.com/endpoints?modals=endpoint-details&model-id=deepseek-ai/DeepSeek-V4.1-Flash",+ "supports_vision": true+ }, "nebius/MiniMaxAI/MiniMax-M2.5": { "max_tokens": 196608, "max_input_tokens": 196608,@@ -56230,7 +56372,7 @@ "input_cost_per_audio_token": 3e-06, "input_cost_per_token": 5e-07, "litellm_provider": "gemini",- "max_input_tokens": 1048576,+ "max_input_tokens": 131072, "max_output_tokens": 8192, "max_tokens": 8192, "mode": "realtime",@@ -56417,7 +56559,7 @@ "input_cost_per_audio_token": 3e-06, "input_cost_per_token": 5e-07, "litellm_provider": "gemini",- "max_input_tokens": 1048576,+ "max_input_tokens": 131072, "max_output_tokens": 8192, "max_tokens": 8192, "mode": "realtime",@@ -59653,14 +59795,18 @@ "supports_anthropic_compaction": true, "cache_creation_input_token_cost": 1.25e-05, "cache_creation_input_token_cost_above_1hr": 2e-05,+ "cache_creation_input_token_cost_batches": 6.25e-06, "cache_read_input_token_cost": 1e-06,+ "cache_read_input_token_cost_batches": 5e-07, "input_cost_per_token": 1e-05,+ "input_cost_per_token_batches": 5e-06, "litellm_provider": "anthropic", "max_input_tokens": 1000000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 5e-05,+ "output_cost_per_token_batches": 2.5e-05, "prompt_cache_min_tokens": 512, "search_context_cost_per_query": { "search_context_size_high": 0.01,@@ -59693,14 +59839,18 @@ "supports_anthropic_compaction": true, "cache_creation_input_token_cost": 1.25e-05, "cache_creation_input_token_cost_above_1hr": 2e-05,+ "cache_creation_input_token_cost_batches": 6.25e-06, "cache_read_input_token_cost": 2.5e-07,+ "cache_read_input_token_cost_batches": 1.25e-07, "input_cost_per_token": 1e-05,+ "input_cost_per_token_batches": 5e-06, "litellm_provider": "anthropic", "max_input_tokens": 1000000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 5e-05,+ "output_cost_per_token_batches": 2.5e-05, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -63000,6 +63150,22 @@ "image" ] },+ "xai/grok-imagine-image-pro": {+ "input_cost_per_image": 0.05,+ "litellm_provider": "xai",+ "mode": "image_generation",+ "source": "https://docs.x.ai/docs/models",+ "supported_endpoints": [+ "/v1/images/generations"+ ],+ "supported_modalities": [+ "text",+ "image"+ ],+ "supported_output_modalities": [+ "image"+ ]+ }, "xai/grok-imagine-image-2.0": { "input_cost_per_image": 0.06, "litellm_provider": "xai",@@ -64006,6 +64172,55 @@ "supports_response_schema": true, "supports_vision": true },+ "azure_ai/deepseek-r1": {+ "input_cost_per_token": 1.35e-06,+ "output_cost_per_token": 5.4e-06,+ "litellm_provider": "azure_ai",+ "mode": "chat",+ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'"+ },+ "azure_ai/deepseek-v3-0324": {+ "input_cost_per_token": 1.14e-06,+ "output_cost_per_token": 4.56e-06,+ "litellm_provider": "azure_ai",+ "mode": "chat",+ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'"+ },+ "azure_ai/deepseek-v3.1": {+ "input_cost_per_token": 1.23e-06,+ "output_cost_per_token": 4.94e-06,+ "litellm_provider": "azure_ai",+ "mode": "chat",+ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'"+ },+ "azure_ai/grok-3": {+ "input_cost_per_token": 3e-06,+ "output_cost_per_token": 1.5e-05,+ "litellm_provider": "azure_ai",+ "mode": "chat",+ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'"+ },+ "azure_ai/grok-3-mini": {+ "input_cost_per_token": 2.5e-07,+ "output_cost_per_token": 1.27e-06,+ "litellm_provider": "azure_ai",+ "mode": "chat",+ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'"+ },+ "azure_ai/grok-4-fast-non-reasoning": {+ "input_cost_per_token": 2e-07,+ "output_cost_per_token": 5e-07,+ "litellm_provider": "azure_ai",+ "mode": "chat",+ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'"+ },+ "azure_ai/grok-4-fast-reasoning": {+ "input_cost_per_token": 2e-07,+ "output_cost_per_token": 5e-07,+ "litellm_provider": "azure_ai",+ "mode": "chat",+ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'"+ }, "bedrock/us-gov-west-1/nvidia.nemotron-nano-3-30b": { "input_cost_per_token": 7.2e-08, "litellm_provider": "bedrock",@@ -64834,6 +65049,131 @@ "supports_response_schema": true, "supports_tool_choice": true },+ "bedrock_mantle/deepseek.v3.1": {+ "input_cost_per_token": 5.8e-07,+ "output_cost_per_token": 1.68e-06,+ "litellm_provider": "bedrock_mantle",+ "max_input_tokens": 128000,+ "max_output_tokens": 8000,+ "max_tokens": 8000,+ "mode": "chat",+ "supported_endpoints": [+ "/v1/chat/completions"+ ],+ "supports_function_calling": true,+ "supports_tool_choice": true,+ "source": "https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-deepseek-deepseek-v3-1.html"+ },+ "bedrock_mantle/moonshotai.kimi-k2-thinking": {+ "input_cost_per_token": 6e-07,+ "output_cost_per_token": 2.5e-06,+ "litellm_provider": "bedrock_mantle",+ "max_input_tokens": 256000,+ "max_output_tokens": 16000,+ "max_tokens": 16000,+ "mode": "chat",+ "supported_endpoints": [+ "/v1/chat/completions"+ ],+ "supports_function_calling": true,+ "supports_tool_choice": true,+ "supports_reasoning": true,+ "source": "https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-moonshot-ai-kimi-k2-thinking.html"+ },+ "bedrock_mantle/qwen.qwen3-235b-a22b-2507": {+ "input_cost_per_token": 2.2e-07,+ "output_cost_per_token": 8.8e-07,+ "litellm_provider": "bedrock_mantle",+ "max_input_tokens": 256000,+ "max_output_tokens": 8000,+ "max_tokens": 8000,+ "mode": "chat",+ "supported_endpoints": [+ "/v1/chat/completions"+ ],+ "supports_function_calling": true,+ "supports_tool_choice": true,+ "supports_reasoning": true,+ "source": "https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-qwen-qwen3-235b-a22b-2507.html"+ },+ "bedrock_mantle/qwen.qwen3-32b": {+ "input_cost_per_token": 1.5e-07,+ "output_cost_per_token": 6e-07,+ "litellm_provider": "bedrock_mantle",+ "max_input_tokens": 32000,+ "max_output_tokens": 8000,+ "max_tokens": 8000,+ "mode": "chat",+ "supported_endpoints": [+ "/v1/chat/completions"+ ],+ "supports_function_calling": true,+ "supports_tool_choice": true,+ "supports_reasoning": true,+ "source": "https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-qwen-qwen3-32b.html"+ },+ "bedrock_mantle/qwen.qwen3-coder-30b-a3b-instruct": {+ "input_cost_per_token": 1.5e-07,+ "output_cost_per_token": 6e-07,+ "litellm_provider": "bedrock_mantle",+ "max_input_tokens": 256000,+ "max_output_tokens": 16000,+ "max_tokens": 16000,+ "mode": "chat",+ "supported_endpoints": [+ "/v1/chat/completions"+ ],+ "supports_function_calling": true,+ "supports_tool_choice": true,+ "source": "https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-qwen-qwen3-coder-30b-a3b-instruct.html"+ },+ "bedrock_mantle/qwen.qwen3-coder-480b-a35b-instruct": {+ "input_cost_per_token": 4.5e-07,+ "output_cost_per_token": 1.8e-06,+ "litellm_provider": "bedrock_mantle",+ "max_input_tokens": 128000,+ "max_output_tokens": 16000,+ "max_tokens": 16000,+ "mode": "chat",+ "supported_endpoints": [+ "/v1/chat/completions"+ ],+ "supports_function_calling": true,+ "supports_tool_choice": true,+ "source": "https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-qwen-qwen3-coder-480b-a35b-instruct.html"+ },+ "bedrock_mantle/qwen.qwen3-next-80b-a3b-instruct": {+ "input_cost_per_token": 1.4e-07,+ "output_cost_per_token": 1.2e-06,+ "litellm_provider": "bedrock_mantle",+ "max_input_tokens": 256000,+ "max_output_tokens": 8000,+ "max_tokens": 8000,+ "mode": "chat",+ "supported_endpoints": [+ "/v1/chat/completions"+ ],+ "supports_function_calling": true,+ "supports_tool_choice": true,+ "supports_reasoning": true,+ "source": "https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-qwen-qwen3-next-80b-a3b.html"+ },+ "bedrock_mantle/qwen.qwen3-vl-235b-a22b-instruct": {+ "input_cost_per_token": 5.3e-07,+ "output_cost_per_token": 2.66e-06,+ "litellm_provider": "bedrock_mantle",+ "max_input_tokens": 256000,+ "max_output_tokens": 8000,+ "max_tokens": 8000,+ "mode": "chat",+ "supported_endpoints": [+ "/v1/chat/completions"+ ],+ "supports_function_calling": true,+ "supports_tool_choice": true,+ "supports_vision": true,+ "source": "https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-qwen-qwen3-vl-235b-a22b.html"+ }, "azure/us-gov/gpt-5.1": { "cache_read_input_token_cost": 1.71875e-07, "default_reasoning_effort": "none",@@ -76578,6 +76918,23 @@ "supports_vision": true, "supports_web_search": false },+ "openrouter/perceptron/perceptron-mk1.5": {+ "input_cost_per_token": 1.5e-07,+ "litellm_provider": "openrouter",+ "max_input_tokens": 36864,+ "max_output_tokens": 8192,+ "max_tokens": 8192,+ "mode": "chat",+ "output_cost_per_token": 1.5e-06,+ "source": "https://openrouter.ai/api/v1/models",+ "supports_audio_input": true,+ "supports_function_calling": true,+ "supports_reasoning": true,+ "supports_response_schema": true,+ "supports_tool_choice": true,+ "supports_video_input": true,+ "supports_vision": true+ }, "vertex_ai/gemini-2.0-flash": { "deprecation_date": "2026-06-01", "input_cost_per_audio_token": 1e-06,litellm/types/utils.pyindex cd336c9b989..2e518af4da4 100644@@ -302,6 +302,7 @@ class ModelInfoBase(ProviderSpecificModelInfo, total=False): cache_read_input_token_cost_above_200k_tokens_batches: ReadOnly[float | None] cache_read_input_token_cost_above_272k_tokens_batches: ReadOnly[float | None] cache_creation_input_token_cost_batches: ReadOnly[float | None]+ cache_creation_input_token_cost_above_200k_tokens_batches: ReadOnly[float | None] cache_creation_input_token_cost_above_272k_tokens_batches: ReadOnly[float | None] # Smallest prefix this model will actually cache, whatever caching mechanism its provider uses. # Absent means the provider-agnostic default applies; see MINIMUM_PROMPT_CACHE_TOKEN_COUNT.@@ -3735,6 +3736,7 @@ class CustomPricingLiteLLMParams(MirroredPricingParams): cache_read_input_token_cost_above_200k_tokens_batches: float | None = None cache_read_input_token_cost_above_272k_tokens_batches: float | None = None cache_creation_input_token_cost_batches: float | None = None+ cache_creation_input_token_cost_above_200k_tokens_batches: float | None = None cache_creation_input_token_cost_above_272k_tokens_batches: float | None = None cache_read_input_audio_token_cost: float | None = None cache_read_input_image_token_cost: float | None = Nonelitellm/utils.pyindex be4388802f9..42da2e2a7b7 100644@@ -6167,6 +6167,9 @@ def _get_model_info_helper( "cache_read_input_token_cost_above_272k_tokens_batches" ), cache_creation_input_token_cost_batches=_model_info.get("cache_creation_input_token_cost_batches"),+ cache_creation_input_token_cost_above_200k_tokens_batches=_model_info.get(+ "cache_creation_input_token_cost_above_200k_tokens_batches"+ ), cache_creation_input_token_cost_above_272k_tokens_batches=_model_info.get( "cache_creation_input_token_cost_above_272k_tokens_batches" ),model_prices_and_context_window.jsonindex 56129fef135..ec31039fee0 100644@@ -1279,8 +1279,11 @@ "source": "https://pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrockFoundationModels/current/index.json" }, "anthropic.claude-mythos-preview": {- "input_cost_per_token": 0,- "output_cost_per_token": 0,+ "cache_creation_input_token_cost": 3.4375e-05,+ "cache_creation_input_token_cost_above_1hr": 5.5e-05,+ "cache_read_input_token_cost": 2.75e-06,+ "input_cost_per_token": 2.75e-05,+ "output_cost_per_token": 0.0001375, "litellm_provider": "bedrock", "max_input_tokens": 1000000, "max_output_tokens": 128000,@@ -1289,10 +1292,11 @@ "thinking_always_on": true, "supports_function_calling": true, "supports_vision": true,- "supports_prompt_caching": false,+ "supports_prompt_caching": true, "supports_reasoning": true, "supports_tool_choice": true,- "supports_output_config": true+ "supports_output_config": true,+ "source": "https://pricing.us-east-1.amazonaws.com/offers/v1.0/aws/AmazonBedrockFoundationModels/current/index.json" }, "global.anthropic.claude-opus-4-7": { "bedrock_converse_supports_strict_tools": false,@@ -7773,27 +7777,27 @@ "output_cost_per_token_above_272k_tokens_batches": 0.000135 }, "azure/gpt-5.6": {- "cache_creation_input_token_cost": 6.25e-06,- "cache_creation_input_token_cost_above_272k_tokens": 1.25e-05,- "cache_creation_input_token_cost_priority": 1.25e-05,- "cache_creation_input_token_cost_above_272k_tokens_priority": 2.5e-05,- "cache_read_input_token_cost": 5e-07,- "cache_read_input_token_cost_above_272k_tokens": 1e-06,- "cache_read_input_token_cost_priority": 1e-06,- "cache_read_input_token_cost_above_272k_tokens_priority": 2e-06,- "input_cost_per_token": 5e-06,- "input_cost_per_token_above_272k_tokens": 1e-05,- "input_cost_per_token_priority": 1e-05,- "input_cost_per_token_above_272k_tokens_priority": 2e-05,+ "cache_creation_input_token_cost": 5e-06,+ "cache_creation_input_token_cost_above_272k_tokens": 1e-05,+ "cache_creation_input_token_cost_priority": 1e-05,+ "cache_creation_input_token_cost_above_272k_tokens_priority": 2e-05,+ "cache_read_input_token_cost": 4e-07,+ "cache_read_input_token_cost_above_272k_tokens": 8e-07,+ "cache_read_input_token_cost_priority": 8e-07,+ "cache_read_input_token_cost_above_272k_tokens_priority": 1.6e-06,+ "input_cost_per_token": 4e-06,+ "input_cost_per_token_above_272k_tokens": 8e-06,+ "input_cost_per_token_priority": 8e-06,+ "input_cost_per_token_above_272k_tokens_priority": 1.6e-05, "litellm_provider": "azure", "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat",- "output_cost_per_token": 3e-05,- "output_cost_per_token_above_272k_tokens": 4.5e-05,- "output_cost_per_token_priority": 6e-05,- "output_cost_per_token_above_272k_tokens_priority": 9e-05,+ "output_cost_per_token": 2e-05,+ "output_cost_per_token_above_272k_tokens": 3e-05,+ "output_cost_per_token_priority": 4e-05,+ "output_cost_per_token_above_272k_tokens_priority": 6e-05, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -8579,27 +8583,27 @@ "supports_web_search": true }, "azure/us/gpt-5.6": {- "cache_creation_input_token_cost": 6.875e-06,- "cache_creation_input_token_cost_above_272k_tokens": 1.375e-05,- "cache_creation_input_token_cost_above_272k_tokens_priority": 2.75e-05,- "cache_creation_input_token_cost_priority": 1.375e-05,- "cache_read_input_token_cost": 5.5e-07,- "cache_read_input_token_cost_above_272k_tokens": 1.1e-06,- "cache_read_input_token_cost_above_272k_tokens_priority": 2.2e-06,- "cache_read_input_token_cost_priority": 1.1e-06,- "input_cost_per_token": 5.5e-06,- "input_cost_per_token_above_272k_tokens": 1.1e-05,- "input_cost_per_token_above_272k_tokens_priority": 2.2e-05,- "input_cost_per_token_priority": 1.1e-05,+ "cache_creation_input_token_cost": 5.5e-06,+ "cache_creation_input_token_cost_above_272k_tokens": 1.1e-05,+ "cache_creation_input_token_cost_above_272k_tokens_priority": 2.2e-05,+ "cache_creation_input_token_cost_priority": 1.1e-05,+ "cache_read_input_token_cost": 4.4e-07,+ "cache_read_input_token_cost_above_272k_tokens": 8.8e-07,+ "cache_read_input_token_cost_above_272k_tokens_priority": 1.76e-06,+ "cache_read_input_token_cost_priority": 8.8e-07,+ "input_cost_per_token": 4.4e-06,+ "input_cost_per_token_above_272k_tokens": 8.8e-06,+ "input_cost_per_token_above_272k_tokens_priority": 1.76e-05,+ "input_cost_per_token_priority": 8.8e-06, "litellm_provider": "azure", "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat",- "output_cost_per_token": 3.3e-05,- "output_cost_per_token_above_272k_tokens": 4.95e-05,- "output_cost_per_token_above_272k_tokens_priority": 9.9e-05,- "output_cost_per_token_priority": 6.6e-05,+ "output_cost_per_token": 2.2e-05,+ "output_cost_per_token_above_272k_tokens": 3.3e-05,+ "output_cost_per_token_above_272k_tokens_priority": 6.6e-05,+ "output_cost_per_token_priority": 4.4e-05, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -8985,27 +8989,27 @@ "supports_web_search": true }, "azure/eu/gpt-5.6": {- "cache_creation_input_token_cost": 6.875e-06,- "cache_creation_input_token_cost_above_272k_tokens": 1.375e-05,- "cache_creation_input_token_cost_above_272k_tokens_priority": 2.75e-05,- "cache_creation_input_token_cost_priority": 1.375e-05,- "cache_read_input_token_cost": 5.5e-07,- "cache_read_input_token_cost_above_272k_tokens": 1.1e-06,- "cache_read_input_token_cost_above_272k_tokens_priority": 2.2e-06,- "cache_read_input_token_cost_priority": 1.1e-06,- "input_cost_per_token": 5.5e-06,- "input_cost_per_token_above_272k_tokens": 1.1e-05,- "input_cost_per_token_above_272k_tokens_priority": 2.2e-05,- "input_cost_per_token_priority": 1.1e-05,+ "cache_creation_input_token_cost": 5.5e-06,+ "cache_creation_input_token_cost_above_272k_tokens": 1.1e-05,+ "cache_creation_input_token_cost_above_272k_tokens_priority": 2.2e-05,+ "cache_creation_input_token_cost_priority": 1.1e-05,+ "cache_read_input_token_cost": 4.4e-07,+ "cache_read_input_token_cost_above_272k_tokens": 8.8e-07,+ "cache_read_input_token_cost_above_272k_tokens_priority": 1.76e-06,+ "cache_read_input_token_cost_priority": 8.8e-07,+ "input_cost_per_token": 4.4e-06,+ "input_cost_per_token_above_272k_tokens": 8.8e-06,+ "input_cost_per_token_above_272k_tokens_priority": 1.76e-05,+ "input_cost_per_token_priority": 8.8e-06, "litellm_provider": "azure", "max_input_tokens": 922000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat",- "output_cost_per_token": 3.3e-05,- "output_cost_per_token_above_272k_tokens": 4.95e-05,- "output_cost_per_token_above_272k_tokens_priority": 9.9e-05,- "output_cost_per_token_priority": 6.6e-05,+ "output_cost_per_token": 2.2e-05,+ "output_cost_per_token_above_272k_tokens": 3.3e-05,+ "output_cost_per_token_above_272k_tokens_priority": 6.6e-05,+ "output_cost_per_token_priority": 4.4e-05, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -14640,14 +14644,18 @@ "claude-haiku-4-5-20251001": { "cache_creation_input_token_cost": 1.25e-06, "cache_creation_input_token_cost_above_1hr": 2e-06,+ "cache_creation_input_token_cost_batches": 6.25e-07, "cache_read_input_token_cost": 1e-07,+ "cache_read_input_token_cost_batches": 5e-08, "input_cost_per_token": 1e-06,+ "input_cost_per_token_batches": 5e-07, "litellm_provider": "anthropic", "max_input_tokens": 200000, "max_output_tokens": 64000, "max_tokens": 64000, "mode": "chat", "output_cost_per_token": 5e-06,+ "output_cost_per_token_batches": 2.5e-06, "supports_assistant_prefill": true, "supports_function_calling": true, "supports_native_structured_output": true,@@ -14663,14 +14671,18 @@ "claude-haiku-4-5": { "cache_creation_input_token_cost": 1.25e-06, "cache_creation_input_token_cost_above_1hr": 2e-06,+ "cache_creation_input_token_cost_batches": 6.25e-07, "cache_read_input_token_cost": 1e-07,+ "cache_read_input_token_cost_batches": 5e-08, "input_cost_per_token": 1e-06,+ "input_cost_per_token_batches": 5e-07, "litellm_provider": "anthropic", "max_input_tokens": 200000, "max_output_tokens": 64000, "max_tokens": 64000, "mode": "chat", "output_cost_per_token": 5e-06,+ "output_cost_per_token_batches": 2.5e-06, "supports_assistant_prefill": true, "supports_function_calling": true, "supports_native_structured_output": true,@@ -14693,13 +14705,21 @@ "input_cost_per_token_above_200k_tokens": 6e-06, "output_cost_per_token_above_200k_tokens": 2.25e-05, "cache_creation_input_token_cost_above_200k_tokens": 7.5e-06,+ "cache_creation_input_token_cost_above_200k_tokens_batches": 3.75e-06,+ "cache_creation_input_token_cost_batches": 1.875e-06, "cache_read_input_token_cost_above_200k_tokens": 6e-07,+ "cache_read_input_token_cost_above_200k_tokens_batches": 3e-07,+ "cache_read_input_token_cost_batches": 1.5e-07,+ "input_cost_per_token_above_200k_tokens_batches": 3e-06,+ "input_cost_per_token_batches": 1.5e-06, "litellm_provider": "anthropic", "max_input_tokens": 1000000, "max_output_tokens": 64000, "max_tokens": 64000, "mode": "chat", "output_cost_per_token": 1.5e-05,+ "output_cost_per_token_above_200k_tokens_batches": 1.125e-05,+ "output_cost_per_token_batches": 7.5e-06, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -14727,13 +14747,21 @@ "input_cost_per_token_above_200k_tokens": 6e-06, "output_cost_per_token_above_200k_tokens": 2.25e-05, "cache_creation_input_token_cost_above_200k_tokens": 7.5e-06,+ "cache_creation_input_token_cost_above_200k_tokens_batches": 3.75e-06,+ "cache_creation_input_token_cost_batches": 1.875e-06, "cache_read_input_token_cost_above_200k_tokens": 6e-07,+ "cache_read_input_token_cost_above_200k_tokens_batches": 3e-07,+ "cache_read_input_token_cost_batches": 1.5e-07,+ "input_cost_per_token_above_200k_tokens_batches": 3e-06,+ "input_cost_per_token_batches": 1.5e-06, "litellm_provider": "anthropic", "max_input_tokens": 1000000, "max_output_tokens": 64000, "max_tokens": 64000, "mode": "chat", "output_cost_per_token": 1.5e-05,+ "output_cost_per_token_above_200k_tokens_batches": 1.125e-05,+ "output_cost_per_token_batches": 7.5e-06, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -14757,14 +14785,18 @@ "supports_anthropic_compaction": true, "cache_creation_input_token_cost": 2.5e-06, "cache_creation_input_token_cost_above_1hr": 4e-06,+ "cache_creation_input_token_cost_batches": 1.25e-06, "cache_read_input_token_cost": 2e-07,+ "cache_read_input_token_cost_batches": 1e-07, "input_cost_per_token": 2e-06,+ "input_cost_per_token_batches": 1e-06, "litellm_provider": "anthropic", "max_input_tokens": 1000000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 1e-05,+ "output_cost_per_token_batches": 5e-06, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -14797,14 +14829,18 @@ "supports_anthropic_compaction": true, "cache_creation_input_token_cost": 3.75e-06, "cache_creation_input_token_cost_above_1hr": 6e-06,+ "cache_creation_input_token_cost_batches": 1.875e-06, "cache_read_input_token_cost": 3e-07,+ "cache_read_input_token_cost_batches": 1.5e-07, "input_cost_per_token": 3e-06,+ "input_cost_per_token_batches": 1.5e-06, "litellm_provider": "anthropic", "max_input_tokens": 1000000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 1.5e-05,+ "output_cost_per_token_batches": 7.5e-06, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -14865,14 +14901,18 @@ "claude-opus-4-5-20251101": { "cache_creation_input_token_cost": 6.25e-06, "cache_creation_input_token_cost_above_1hr": 1e-05,+ "cache_creation_input_token_cost_batches": 3.125e-06, "cache_read_input_token_cost": 5e-07,+ "cache_read_input_token_cost_batches": 2.5e-07, "input_cost_per_token": 5e-06,+ "input_cost_per_token_batches": 2.5e-06, "litellm_provider": "anthropic", "max_input_tokens": 200000, "max_output_tokens": 64000, "max_tokens": 64000, "mode": "chat", "output_cost_per_token": 2.5e-05,+ "output_cost_per_token_batches": 1.25e-05, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -14895,14 +14935,18 @@ "claude-opus-4-5": { "cache_creation_input_token_cost": 6.25e-06, "cache_creation_input_token_cost_above_1hr": 1e-05,+ "cache_creation_input_token_cost_batches": 3.125e-06, "cache_read_input_token_cost": 5e-07,+ "cache_read_input_token_cost_batches": 2.5e-07, "input_cost_per_token": 5e-06,+ "input_cost_per_token_batches": 2.5e-06, "litellm_provider": "anthropic", "max_input_tokens": 200000, "max_output_tokens": 64000, "max_tokens": 64000, "mode": "chat", "output_cost_per_token": 2.5e-05,+ "output_cost_per_token_batches": 1.25e-05, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -14927,14 +14971,18 @@ "supports_anthropic_compaction": true, "cache_creation_input_token_cost": 6.25e-06, "cache_creation_input_token_cost_above_1hr": 1e-05,+ "cache_creation_input_token_cost_batches": 3.125e-06, "cache_read_input_token_cost": 5e-07,+ "cache_read_input_token_cost_batches": 2.5e-07, "input_cost_per_token": 5e-06,+ "input_cost_per_token_batches": 2.5e-06, "litellm_provider": "anthropic", "max_input_tokens": 1000000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 2.5e-05,+ "output_cost_per_token_batches": 1.25e-05, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -14966,14 +15014,18 @@ "supports_anthropic_compaction": true, "cache_creation_input_token_cost": 6.25e-06, "cache_creation_input_token_cost_above_1hr": 1e-05,+ "cache_creation_input_token_cost_batches": 3.125e-06, "cache_read_input_token_cost": 5e-07,+ "cache_read_input_token_cost_batches": 2.5e-07, "input_cost_per_token": 5e-06,+ "input_cost_per_token_batches": 2.5e-06, "litellm_provider": "anthropic", "max_input_tokens": 1000000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 2.5e-05,+ "output_cost_per_token_batches": 1.25e-05, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -15004,14 +15056,18 @@ "supports_anthropic_compaction": true, "cache_creation_input_token_cost": 6.25e-06, "cache_creation_input_token_cost_above_1hr": 1e-05,+ "cache_creation_input_token_cost_batches": 3.125e-06, "cache_read_input_token_cost": 5e-07,+ "cache_read_input_token_cost_batches": 2.5e-07, "input_cost_per_token": 5e-06,+ "input_cost_per_token_batches": 2.5e-06, "litellm_provider": "anthropic", "max_input_tokens": 1000000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 2.5e-05,+ "output_cost_per_token_batches": 1.25e-05, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -15044,14 +15100,18 @@ "supports_anthropic_compaction": true, "cache_creation_input_token_cost": 6.25e-06, "cache_creation_input_token_cost_above_1hr": 1e-05,+ "cache_creation_input_token_cost_batches": 3.125e-06, "cache_read_input_token_cost": 5e-07,+ "cache_read_input_token_cost_batches": 2.5e-07, "input_cost_per_token": 5e-06,+ "input_cost_per_token_batches": 2.5e-06, "litellm_provider": "anthropic", "max_input_tokens": 1000000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 2.5e-05,+ "output_cost_per_token_batches": 1.25e-05, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -15083,14 +15143,18 @@ "supports_anthropic_compaction": true, "cache_creation_input_token_cost": 1.25e-05, "cache_creation_input_token_cost_above_1hr": 2e-05,+ "cache_creation_input_token_cost_batches": 6.25e-06, "cache_read_input_token_cost": 1e-06,+ "cache_read_input_token_cost_batches": 5e-07, "input_cost_per_token": 1e-05,+ "input_cost_per_token_batches": 5e-06, "litellm_provider": "anthropic", "max_input_tokens": 1000000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 5e-05,+ "output_cost_per_token_batches": 2.5e-05, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -15123,14 +15187,18 @@ "supports_anthropic_compaction": true, "cache_creation_input_token_cost": 1.25e-05, "cache_creation_input_token_cost_above_1hr": 2e-05,+ "cache_creation_input_token_cost_batches": 6.25e-06, "cache_read_input_token_cost": 2.5e-07,+ "cache_read_input_token_cost_batches": 1.25e-07, "input_cost_per_token": 1e-05,+ "input_cost_per_token_batches": 5e-06, "litellm_provider": "anthropic", "max_input_tokens": 1000000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 5e-05,+ "output_cost_per_token_batches": 2.5e-05, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -15164,14 +15232,18 @@ "supports_anthropic_compaction": true, "cache_creation_input_token_cost": 5e-06, "cache_creation_input_token_cost_above_1hr": 8e-06,+ "cache_creation_input_token_cost_batches": 2.5e-06, "cache_read_input_token_cost": 2e-07,+ "cache_read_input_token_cost_batches": 1e-07, "input_cost_per_token": 4e-06,+ "input_cost_per_token_batches": 2e-06, "litellm_provider": "anthropic", "max_input_tokens": 1000000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 2e-05,+ "output_cost_per_token_batches": 1e-05, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -15207,14 +15279,18 @@ "supports_anthropic_compaction": true, "cache_creation_input_token_cost": 6.25e-06, "cache_creation_input_token_cost_above_1hr": 1e-05,+ "cache_creation_input_token_cost_batches": 3.125e-06, "cache_read_input_token_cost": 5e-07,+ "cache_read_input_token_cost_batches": 2.5e-07, "input_cost_per_token": 5e-06,+ "input_cost_per_token_batches": 2.5e-06, "litellm_provider": "anthropic", "max_input_tokens": 1000000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 2.5e-05,+ "output_cost_per_token_batches": 1.25e-05, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -15250,14 +15326,18 @@ "supports_anthropic_compaction": true, "cache_creation_input_token_cost": 6.25e-06, "cache_creation_input_token_cost_above_1hr": 1e-05,+ "cache_creation_input_token_cost_batches": 3.125e-06, "cache_read_input_token_cost": 5e-07,+ "cache_read_input_token_cost_batches": 2.5e-07, "input_cost_per_token": 5e-06,+ "input_cost_per_token_batches": 2.5e-06, "litellm_provider": "anthropic", "max_input_tokens": 1000000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 2.5e-05,+ "output_cost_per_token_batches": 1.25e-05, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -15756,6 +15836,16 @@ "supports_function_calling": true, "supports_tool_choice": true },+ "c4ai-aya-expanse-32b": {+ "input_cost_per_token": 5e-07,+ "output_cost_per_token": 1.5e-06,+ "litellm_provider": "cohere_chat",+ "max_input_tokens": 128000,+ "max_output_tokens": 4000,+ "max_tokens": 4000,+ "mode": "chat",+ "source": "https://docs.cohere.com/docs/models"+ }, "command-a-plus-05-2026": { "input_cost_per_token": 0.0, "litellm_provider": "cohere_chat",@@ -22145,11 +22235,11 @@ } }, "bing_grounding/search": {- "input_cost_per_query": 0.035,+ "input_cost_per_query": 0.014, "litellm_provider": "bing_grounding", "mode": "search", "metadata": {- "notes": "Grounding with Bing Search (G1 SKU): $35 per 1,000 transactions. Tokens for the Foundry model deployment that runs the grounded search are billed separately on that deployment."+ "notes": "Grounding with Bing Search (G1 SKU): $14 per 1,000 transactions. Tokens for the Foundry model deployment that runs the grounded search are billed separately on that deployment." } }, "tinyfish/search": {@@ -28036,6 +28126,7 @@ "tpm": 10000000 }, "gemini/gemini-3-pro-image-preview": {+ "deprecation_date": "2026-06-25", "input_cost_per_image": 0.0011, "input_cost_per_token": 2e-06, "input_cost_per_token_batches": 1e-06,@@ -28083,6 +28174,7 @@ "supports_reasoning": false }, "gemini/gemini-3.1-flash-image-preview": {+ "deprecation_date": "2026-06-25", "input_cost_per_token": 5e-07, "input_cost_per_token_batches": 2.5e-07, "litellm_provider": "gemini",@@ -28130,6 +28222,7 @@ "cache_read_input_token_cost_batches": 1.25e-08, "cache_read_input_token_cost_flex": 1.25e-08, "cache_read_input_token_cost_priority": 4.5e-08,+ "deprecation_date": "2026-05-25", "input_cost_per_audio_token": 5e-07, "input_cost_per_token": 2.5e-07, "input_cost_per_token_batches": 1.25e-07,@@ -30551,7 +30644,11 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true,+ "supports_none_reasoning_effort": true,+ "supports_xhigh_reasoning_effort": true,+ "supports_minimal_reasoning_effort": false }, "chatgpt/gpt-5.6-luna": { "litellm_provider": "chatgpt",@@ -30567,7 +30664,11 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_minimal_reasoning_effort": false,+ "supports_none_reasoning_effort": true,+ "supports_reasoning": true,+ "supports_xhigh_reasoning_effort": true }, "chatgpt/gpt-5.6-sol": { "litellm_provider": "chatgpt",@@ -30583,7 +30684,11 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_minimal_reasoning_effort": false,+ "supports_none_reasoning_effort": true,+ "supports_reasoning": true,+ "supports_xhigh_reasoning_effort": true }, "chatgpt/gpt-5.6-terra": { "litellm_provider": "chatgpt",@@ -30599,7 +30704,11 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_minimal_reasoning_effort": false,+ "supports_none_reasoning_effort": true,+ "supports_reasoning": true,+ "supports_xhigh_reasoning_effort": true }, "chatgpt/gpt-5.4": { "litellm_provider": "chatgpt",@@ -30614,7 +30723,12 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true,+ "supports_none_reasoning_effort": true,+ "default_reasoning_effort": "none",+ "supports_xhigh_reasoning_effort": true,+ "supports_minimal_reasoning_effort": false }, "chatgpt/gpt-5.4-pro": { "litellm_provider": "chatgpt",@@ -30628,7 +30742,11 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true,+ "supports_none_reasoning_effort": false,+ "supports_xhigh_reasoning_effort": true,+ "supports_minimal_reasoning_effort": false }, "chatgpt/gpt-5.3-codex": { "litellm_provider": "chatgpt",@@ -30642,7 +30760,11 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true,+ "supports_none_reasoning_effort": false,+ "supports_xhigh_reasoning_effort": false,+ "supports_minimal_reasoning_effort": true }, "chatgpt/gpt-5.3-codex-spark": { "litellm_provider": "chatgpt",@@ -30686,7 +30808,8 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true }, "chatgpt/gpt-5.2-codex": { "litellm_provider": "chatgpt",@@ -30700,7 +30823,8 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true }, "chatgpt/gpt-5.2": { "litellm_provider": "chatgpt",@@ -30715,7 +30839,12 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true,+ "supports_none_reasoning_effort": true,+ "default_reasoning_effort": "none",+ "supports_xhigh_reasoning_effort": true,+ "supports_minimal_reasoning_effort": false }, "chatgpt/gpt-5.1-codex-max": { "litellm_provider": "chatgpt",@@ -30729,7 +30858,8 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true }, "chatgpt/gpt-5.1-codex-mini": { "litellm_provider": "chatgpt",@@ -30743,7 +30873,8 @@ "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_response_schema": true,- "supports_vision": true+ "supports_vision": true,+ "supports_reasoning": true }, "gigachat/GigaChat-2": { "input_cost_per_token": 0.0,@@ -39214,6 +39345,17 @@ "supports_function_calling": true, "supports_reasoning": true },+ "nebius/deepseek-ai/DeepSeek-V4.1-Flash": {+ "input_cost_per_token": 3e-07,+ "litellm_provider": "nebius",+ "max_input_tokens": 1048576,+ "max_output_tokens": 1048576,+ "max_tokens": 1048576,+ "mode": "chat",+ "output_cost_per_token": 1.2e-06,+ "source": "https://tokenfactory.nebius.com/endpoints?modals=endpoint-details&model-id=deepseek-ai/DeepSeek-V4.1-Flash",+ "supports_vision": true+ }, "nebius/MiniMaxAI/MiniMax-M2.5": { "max_tokens": 196608, "max_input_tokens": 196608,@@ -56230,7 +56372,7 @@ "input_cost_per_audio_token": 3e-06, "input_cost_per_token": 5e-07, "litellm_provider": "gemini",- "max_input_tokens": 1048576,+ "max_input_tokens": 131072, "max_output_tokens": 8192, "max_tokens": 8192, "mode": "realtime",@@ -56417,7 +56559,7 @@ "input_cost_per_audio_token": 3e-06, "input_cost_per_token": 5e-07, "litellm_provider": "gemini",- "max_input_tokens": 1048576,+ "max_input_tokens": 131072, "max_output_tokens": 8192, "max_tokens": 8192, "mode": "realtime",@@ -59653,14 +59795,18 @@ "supports_anthropic_compaction": true, "cache_creation_input_token_cost": 1.25e-05, "cache_creation_input_token_cost_above_1hr": 2e-05,+ "cache_creation_input_token_cost_batches": 6.25e-06, "cache_read_input_token_cost": 1e-06,+ "cache_read_input_token_cost_batches": 5e-07, "input_cost_per_token": 1e-05,+ "input_cost_per_token_batches": 5e-06, "litellm_provider": "anthropic", "max_input_tokens": 1000000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 5e-05,+ "output_cost_per_token_batches": 2.5e-05, "prompt_cache_min_tokens": 512, "search_context_cost_per_query": { "search_context_size_high": 0.01,@@ -59693,14 +59839,18 @@ "supports_anthropic_compaction": true, "cache_creation_input_token_cost": 1.25e-05, "cache_creation_input_token_cost_above_1hr": 2e-05,+ "cache_creation_input_token_cost_batches": 6.25e-06, "cache_read_input_token_cost": 2.5e-07,+ "cache_read_input_token_cost_batches": 1.25e-07, "input_cost_per_token": 1e-05,+ "input_cost_per_token_batches": 5e-06, "litellm_provider": "anthropic", "max_input_tokens": 1000000, "max_output_tokens": 128000, "max_tokens": 128000, "mode": "chat", "output_cost_per_token": 5e-05,+ "output_cost_per_token_batches": 2.5e-05, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01,@@ -63000,6 +63150,22 @@ "image" ] },+ "xai/grok-imagine-image-pro": {+ "input_cost_per_image": 0.05,+ "litellm_provider": "xai",+ "mode": "image_generation",+ "source": "https://docs.x.ai/docs/models",+ "supported_endpoints": [+ "/v1/images/generations"+ ],+ "supported_modalities": [+ "text",+ "image"+ ],+ "supported_output_modalities": [+ "image"+ ]+ }, "xai/grok-imagine-image-2.0": { "input_cost_per_image": 0.06, "litellm_provider": "xai",@@ -64006,6 +64172,55 @@ "supports_response_schema": true, "supports_vision": true },+ "azure_ai/deepseek-r1": {+ "input_cost_per_token": 1.35e-06,+ "output_cost_per_token": 5.4e-06,+ "litellm_provider": "azure_ai",+ "mode": "chat",+ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'"+ },+ "azure_ai/deepseek-v3-0324": {+ "input_cost_per_token": 1.14e-06,+ "output_cost_per_token": 4.56e-06,+ "litellm_provider": "azure_ai",+ "mode": "chat",+ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'"+ },+ "azure_ai/deepseek-v3.1": {+ "input_cost_per_token": 1.23e-06,+ "output_cost_per_token": 4.94e-06,+ "litellm_provider": "azure_ai",+ "mode": "chat",+ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'"+ },+ "azure_ai/grok-3": {+ "input_cost_per_token": 3e-06,+ "output_cost_per_token": 1.5e-05,+ "litellm_provider": "azure_ai",+ "mode": "chat",+ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'"+ },+ "azure_ai/grok-3-mini": {+ "input_cost_per_token": 2.5e-07,+ "output_cost_per_token": 1.27e-06,+ "litellm_provider": "azure_ai",+ "mode": "chat",+ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'"+ },+ "azure_ai/grok-4-fast-non-reasoning": {+ "input_cost_per_token": 2e-07,+ "output_cost_per_token": 5e-07,+ "litellm_provider": "azure_ai",+ "mode": "chat",+ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'"+ },+ "azure_ai/grok-4-fast-reasoning": {+ "input_cost_per_token": 2e-07,+ "output_cost_per_token": 5e-07,+ "litellm_provider": "azure_ai",+ "mode": "chat",+ "source": "https://prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20'Foundry%20Models'%20and%20armRegionName%20eq%20'eastus'%20and%20priceType%20eq%20'Consumption'"+ }, "bedrock/us-gov-west-1/nvidia.nemotron-nano-3-30b": { "input_cost_per_token": 7.2e-08, "litellm_provider": "bedrock",@@ -64834,6 +65049,131 @@ "supports_response_schema": true, "supports_tool_choice": true },+ "bedrock_mantle/deepseek.v3.1": {+ "input_cost_per_token": 5.8e-07,+ "output_cost_per_token": 1.68e-06,+ "litellm_provider": "bedrock_mantle",+ "max_input_tokens": 128000,+ "max_output_tokens": 8000,+ "max_tokens": 8000,+ "mode": "chat",+ "supported_endpoints": [+ "/v1/chat/completions"+ ],+ "supports_function_calling": true,+ "supports_tool_choice": true,+ "source": "https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-deepseek-deepseek-v3-1.html"+ },+ "bedrock_mantle/moonshotai.kimi-k2-thinking": {+ "input_cost_per_token": 6e-07,+ "output_cost_per_token": 2.5e-06,+ "litellm_provider": "bedrock_mantle",+ "max_input_tokens": 256000,+ "max_output_tokens": 16000,+ "max_tokens": 16000,+ "mode": "chat",+ "supported_endpoints": [+ "/v1/chat/completions"+ ],+ "supports_function_calling": true,+ "supports_tool_choice": true,+ "supports_reasoning": true,+ "source": "https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-moonshot-ai-kimi-k2-thinking.html"+ },+ "bedrock_mantle/qwen.qwen3-235b-a22b-2507": {+ "input_cost_per_token": 2.2e-07,+ "output_cost_per_token": 8.8e-07,+ "litellm_provider": "bedrock_mantle",+ "max_input_tokens": 256000,+ "max_output_tokens": 8000,+ "max_tokens": 8000,+ "mode": "chat",+ "supported_endpoints": [+ "/v1/chat/completions"+ ],+ "supports_function_calling": true,+ "supports_tool_choice": true,+ "supports_reasoning": true,+ "source": "https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-qwen-qwen3-235b-a22b-2507.html"+ },+ "bedrock_mantle/qwen.qwen3-32b": {+ "input_cost_per_token": 1.5e-07,+ "output_cost_per_token": 6e-07,+ "litellm_provider": "bedrock_mantle",+ "max_input_tokens": 32000,+ "max_output_tokens": 8000,+ "max_tokens": 8000,+ "mode": "chat",+ "supported_endpoints": [+ "/v1/chat/completions"+ ],+ "supports_function_calling": true,+ "supports_tool_choice": true,+ "supports_reasoning": true,+ "source": "https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-qwen-qwen3-32b.html"+ },+ "bedrock_mantle/qwen.qwen3-coder-30b-a3b-instruct": {+ "input_cost_per_token": 1.5e-07,+ "output_cost_per_token": 6e-07,+ "litellm_provider": "bedrock_mantle",+ "max_input_tokens": 256000,+ "max_output_tokens": 16000,+ "max_tokens": 16000,+ "mode": "chat",+ "supported_endpoints": [+ "/v1/chat/completions"+ ],+ "supports_function_calling": true,+ "supports_tool_choice": true,+ "source": "https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-qwen-qwen3-coder-30b-a3b-instruct.html"+ },+ "bedrock_mantle/qwen.qwen3-coder-480b-a35b-instruct": {+ "input_cost_per_token": 4.5e-07,+ "output_cost_per_token": 1.8e-06,+ "litellm_provider": "bedrock_mantle",+ "max_input_tokens": 128000,+ "max_output_tokens": 16000,+ "max_tokens": 16000,+ "mode": "chat",+ "supported_endpoints": [+ "/v1/chat/completions"+ ],+ "supports_function_calling": true,+ "supports_tool_choice": true,+ "source": "https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-qwen-qwen3-coder-480b-a35b-instruct.html"+ },+ "bedrock_mantle/qwen.qwen3-next-80b-a3b-instruct": {+ "input_cost_per_token": 1.4e-07,+ "output_cost_per_token": 1.2e-06,+ "litellm_provider": "bedrock_mantle",+ "max_input_tokens": 256000,+ "max_output_tokens": 8000,+ "max_tokens": 8000,+ "mode": "chat",+ "supported_endpoints": [+ "/v1/chat/completions"+ ],+ "supports_function_calling": true,+ "supports_tool_choice": true,+ "supports_reasoning": true,+ "source": "https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-qwen-qwen3-next-80b-a3b.html"+ },+ "bedrock_mantle/qwen.qwen3-vl-235b-a22b-instruct": {+ "input_cost_per_token": 5.3e-07,+ "output_cost_per_token": 2.66e-06,+ "litellm_provider": "bedrock_mantle",+ "max_input_tokens": 256000,+ "max_output_tokens": 8000,+ "max_tokens": 8000,+ "mode": "chat",+ "supported_endpoints": [+ "/v1/chat/completions"+ ],+ "supports_function_calling": true,+ "supports_tool_choice": true,+ "supports_vision": true,+ "source": "https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-qwen-qwen3-vl-235b-a22b.html"+ }, "azure/us-gov/gpt-5.1": { "cache_read_input_token_cost": 1.71875e-07, "default_reasoning_effort": "none",@@ -76578,6 +76918,23 @@ "supports_vision": true, "supports_web_search": false },+ "openrouter/perceptron/perceptron-mk1.5": {+ "input_cost_per_token": 1.5e-07,+ "litellm_provider": "openrouter",+ "max_input_tokens": 36864,+ "max_output_tokens": 8192,+ "max_tokens": 8192,+ "mode": "chat",+ "output_cost_per_token": 1.5e-06,+ "source": "https://openrouter.ai/api/v1/models",+ "supports_audio_input": true,+ "supports_function_calling": true,+ "supports_reasoning": true,+ "supports_response_schema": true,+ "supports_tool_choice": true,+ "supports_video_input": true,+ "supports_vision": true+ }, "vertex_ai/gemini-2.0-flash": { "deprecation_date": "2026-06-01", "input_cost_per_audio_token": 1e-06,model_prices_and_context_window.schema.jsonindex fa1828c780a..c4048cac905 100644@@ -103,6 +103,11 @@ "minimum": 0, "description": "Rate applied once the prompt exceeds the token threshold in the field name." },+ "cache_creation_input_token_cost_above_200k_tokens_batches": {+ "type": "number",+ "minimum": 0,+ "description": "Rate applied once the prompt exceeds the token threshold in the field name."+ }, "cache_creation_input_token_cost_above_256k_tokens": { "type": "number", "minimum": 0,tests/local_testing/test_get_model_info.pyindex dbe1fc3b69b..79f6739a423 100644@@ -320,6 +320,33 @@ def test_get_model_info_bedrock_cross_region_capability_parity(): assert checked > 0, "no cross-region bedrock profiles found - the filter is inert" ++def test_get_model_info_bedrock_priced_cross_region_profile_has_priced_base():+ os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"+ litellm.model_cost = litellm.get_model_cost_map(url="")++ prefixes = ("us.", "eu.", "apac.", "us-gov.", "au.", "global.")+ checked = 0++ for k, v in litellm.model_cost.items():+ if not str(v.get("litellm_provider", "")).startswith("bedrock"):+ continue+ base_model_key = next(+ (k[len(p) :] for p in prefixes if k.startswith(p)),+ None,+ )+ if base_model_key is None or base_model_key not in litellm.model_cost:+ continue+ checked += 1+ base = litellm.model_cost[base_model_key]+ for cost_key in ("input_cost_per_token", "output_cost_per_token"):+ if (v.get(cost_key) or 0) > 0:+ assert (+ base.get(cost_key) or 0+ ) > 0, f"{k} charges {cost_key} but its base {base_model_key} is free"++ assert checked > 0, "no cross-region bedrock profiles found - the filter is inert"+ def test_get_model_info_huggingface_models(monkeypatch): from litellm import Router from litellm.types.router import ModelGroupInfotests/search_tests/test_bing_grounding_search.pyindex f532158e462..6ea79076370 100644@@ -196,4 +196,5 @@ class TestBingGroundingSearchTransformation: ): response = litellm.search(query="pricing check", search_provider="bing_grounding") - assert response._hidden_params["response_cost"] == pytest.approx(0.035)+ # Grounding with Bing Search (G1 SKU): $14 per 1,000 transactions, https://www.microsoft.com/en-us/bing/apis, checked 2026-09-24+ assert response._hidden_params["response_cost"] == pytest.approx(0.014)tests/unit/litellm_core_utils/llm_cost_calc/test_llm_cost_calc_utils.pyindex 781a3a7c4ed..0afd989272e 100644@@ -3789,3 +3789,15 @@ def test_azure_gpt_6_foundry_price_sheet(_local_model_cost_map, model_base): assert azure_ai_info[field] == base assert azure_us_info[field] == pytest.approx(1.1 * base) assert azure_eu_info[field] == pytest.approx(1.2 * base)+++@pytest.mark.parametrize("region_prefix", ["azure/", "azure/us/", "azure/eu/"])+def test_azure_gpt_5_6_alias_matches_sol_pricing(_local_model_cost_map, region_prefix):+ """The bare gpt-5.6 alias routes to GPT-5.6 Sol, so every Azure region must bill the+ alias exactly like the Sol entry (including the Sept 2026 $4/$20 promo)."""+ alias = litellm.model_cost[f"{region_prefix}gpt-5.6"]+ sol = litellm.model_cost[f"{region_prefix}gpt-5.6-sol"]+ shared_cost_fields = [f for f in alias if "cost" in f and f in sol and not isinstance(alias[f], dict)]+ assert shared_cost_fields+ for field in shared_cost_fields:+ assert alias[field] == sol[field], fieldtests/unit/litellm_core_utils/test_litellm_logging.pyindex d717718cba2..cb1e281e356 100644@@ -7969,6 +7969,29 @@ def test_deployment_pricing_model_info_honors_a_tier_only_batch_override_over_th assert {key: info[key] for key in carried_keys} == {key: _PUBLISHED_BATCH_RATES[key] for key in carried_keys} +@pytest.mark.parametrize(+ "override_key",+ (+ "output_cost_per_token_above_200k_tokens_batches",+ "cache_read_input_token_cost_above_200k_tokens_batches",+ "cache_creation_input_token_cost_above_200k_tokens_batches",+ ),+)+def test_deployment_pricing_model_info_honors_a_200k_tier_batch_override(+ _published_batch_model: None, override_key: str+) -> None:+ from litellm.litellm_core_utils.litellm_logging import deployment_pricing_model_info++ info: Final = deployment_pricing_model_info(_batch_deployment_id({override_key: 1e-3}), _PUBLISHED_BATCH_DEPLOYMENT)+ carried_keys: Final = tuple(+ key for key in (*_PUBLISHED_INPUT_BATCH_KEYS, *_PUBLISHED_OUTPUT_BATCH_KEYS) if key != override_key+ )++ assert info is not None+ assert info[override_key] == 1e-3+ assert {key: info[key] for key in carried_keys} == {key: _PUBLISHED_BATCH_RATES[key] for key in carried_keys}++ def test_get_status_fields_ranks_guardrail_flagged_between_success_and_intervened(): """LIT-6894: a non-blocking flagged verdict must outrank success in the request-level guardrail_status but never mask an intervention."""tests/unit/test_model_prices_schema.pyindex 052278631e2..a05c345b5cd 100644@@ -266,6 +266,61 @@ def test_openai_reasoning_family_entries_carry_supports_reasoning(prices: dict): ) +_ABSENT: Final = object()++REASONING_ANNOTATION_KEYS: Final = (+ "supports_reasoning",+ "supports_minimal_reasoning_effort",+ "supports_none_reasoning_effort",+ "supports_xhigh_reasoning_effort",+ "default_reasoning_effort",+)+++def chatgpt_openai_twins(prices: dict) -> list[tuple[str, str]]:+ """`chatgpt/<model>` rows paired with the bare `<model>` row served by the openai provider.++ Scoped to openai twins on purpose. `ChatGPTConfig` and `ChatGPTResponsesAPIConfig` subclass+ their openai counterparts, so a chatgpt row's reasoning behaviour is whatever the openai row+ describes. The azure rows are a separate registry that already diverges from openai here, and+ pinning them to each other would assert something this repository does not control.+ """+ pairs = []+ for name, entry in prices.items():+ if not isinstance(entry, dict) or not name.startswith("chatgpt/"):+ continue+ bare = name.split("/", 1)[1]+ twin = prices.get(bare)+ if isinstance(twin, dict) and twin.get("litellm_provider") == "openai":+ pairs.append((name, bare))+ return pairs+++def test_chatgpt_rows_carry_their_openai_twin_reasoning_annotations(prices: dict):+ """A chatgpt row must not silently drop the reasoning annotations of the model it proxies.++ `litellm.utils._get_model_info_from_generalization` refuses to fall back when an exact cost-map+ key exists, so an unannotated `chatgpt/<model>` row wins over its annotated twin and+ `/model/info` reports the model as non-reasoning.+ """+ twins = chatgpt_openai_twins(prices)+ assert twins, "no chatgpt/* row has an openai twin any more; this guard has stopped guarding"++ mismatched = []+ for name, bare in twins:+ for key in REASONING_ANNOTATION_KEYS:+ if prices[name].get(key, _ABSENT) != prices[bare].get(key, _ABSENT):+ mismatched.append(+ f"{name}.{key} is {prices[name].get(key)!r}, {bare}.{key} is {prices[bare].get(key)!r}"+ )++ assert mismatched == [], (+ "chatgpt/* entries proxy their openai twin through ChatGPTConfig, so they must carry the "+ "same reasoning annotations; an exact cost-map key blocks the generalization fallback, so "+ "a missing flag here is reported to callers as 'not a reasoning model':\n" + "\n".join(mismatched)+ )++ def test_chat_latest_declares_the_one_effort_openai_accepts(prices: dict): """OpenAI rejects every reasoning.effort on chat-latest except medium, and a reasoning entry with no declared levels resolves to None, which lets /model_group/info and the dashboard offertests/unit/test_utils.pyindex 0cdc52c9a93..5b327305e31 100644@@ -766,6 +766,7 @@ def test_aaamodel_prices_and_context_window_json_is_valid(): "cache_creation_input_token_cost_above_272k_tokens": {"type": "number"}, "cache_creation_input_token_cost_above_272k_tokens_flex": {"type": "number"}, "cache_creation_input_token_cost_above_272k_tokens_priority": {"type": "number"},+ "cache_creation_input_token_cost_above_200k_tokens_batches": {"type": "number"}, "cache_creation_input_token_cost_above_272k_tokens_batches": {"type": "number"}, "cache_creation_input_token_cost_batches": {"type": "number"}, "cache_creation_input_token_cost_flex": {"type": "number"},ui/litellm-dashboard/src/lib/http/schema.d.tsindex 1e0aed46923..6d46af04730 100644@@ -32957,6 +32957,8 @@ export interface components { cache_creation_input_token_cost_above_1hr?: number | null; /** Cache Creation Input Token Cost Above 200K Tokens */ cache_creation_input_token_cost_above_200k_tokens?: number | null;+ /** Cache Creation Input Token Cost Above 200K Tokens Batches */+ cache_creation_input_token_cost_above_200k_tokens_batches?: number | null; /** Cache Creation Input Token Cost Above 272K Tokens */ cache_creation_input_token_cost_above_272k_tokens?: number | null; /** Cache Creation Input Token Cost Above 272K Tokens Batches */@@ -46750,6 +46752,8 @@ export interface components { cache_creation_input_token_cost_above_1hr?: number | null; /** Cache Creation Input Token Cost Above 200K Tokens */ cache_creation_input_token_cost_above_200k_tokens?: number | null;+ /** Cache Creation Input Token Cost Above 200K Tokens Batches */+ cache_creation_input_token_cost_above_200k_tokens_batches?: number | null; /** Cache Creation Input Token Cost Above 272K Tokens */ cache_creation_input_token_cost_above_272k_tokens?: number | null; /** Cache Creation Input Token Cost Above 272K Tokens Batches */