From df78b16941697991a736fc224182fa17270420ed Mon Sep 17 00:00:00 2001 From: Kerry Lu Date: Wed, 4 Nov 2026 08:57:42 -0800 Subject: [PATCH] feat(openai): bill gov.api.openai.com FedRAMP endpoint with 1.1x uplift (#49951) * feat(openai): bill gov.api.openai.com FedRAMP endpoint with 1.1x uplift * chore(openai): trim FedRAMP uplift diff * refactor(openai): bill the FedRAMP endpoint through its own fedramp_uplift_multiplier key * fix(ui): regenerate fedramp_uplift_multiplier doc comment in schema.d.ts * style: ruff format utils.py and reorder fedramp_uplift_multiplier in schema.d.ts * fix(openai): infer data_residency from OPENAI_BASE_URL and litellm.api_base when api_base is unset * test(cost): cover get_fedramp_uplift_multiplier fallback paths * fix(cost): apply fedramp_uplift_multiplier in batch_cost_calculator * fix(openai): derive data_residency from a supplied client's base_url * test: type the supplied-client data_residency test params --------- Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com> --- litellm/cost_calculator.py | 5 +- .../litellm_core_utils/llm_cost_calc/utils.py | 41 +++++-- litellm/llms/openai/cost_calculation.py | 6 +- litellm/llms/openai/data_residency.py | 21 +++- litellm/main.py | 7 +- ...odel_prices_and_context_window_backup.json | 112 ++++++++++++++++++ litellm/types/utils.py | 8 +- litellm/utils.py | 1 + model_prices_and_context_window.json | 112 ++++++++++++++++++ tests/unit/batches/test_batch_utils.py | 4 +- .../llm_cost_calc/test_llm_cost_calc_utils.py | 62 ++++++++++ .../test_get_litellm_params.py | 7 ++ .../test_openai_responses_data_residency.py | 22 ++++ tests/unit/llms/openai/test_data_residency.py | 31 ++++- tests/unit/test_main.py | 25 ++++ tests/unit/test_utils.py | 1 + ui/litellm-dashboard/src/lib/http/schema.d.ts | 4 + 17 files changed, 445 insertions(+), 24 deletions(-) diff --git a/litellm/cost_calculator.py b/litellm/cost_calculator.py index d007e555521..9154a6fdc30 100644 --- a/litellm/cost_calculator.py +++ b/litellm/cost_calculator.py @@ -50,6 +50,7 @@ from litellm.litellm_core_utils.llm_cost_calc.utils import ( get_batch_cost_rates, get_billable_input_tokens, get_cost_per_unit, + get_fedramp_uplift_multiplier, get_modality_token_cost_key, get_regional_uplift_multiplier, get_service_tier_cost_key, @@ -3311,7 +3312,9 @@ def batch_cost_calculator( text_tokens: Final = usage.completion_tokens - image_tokens total_completion_cost = text_tokens * text_rate + image_tokens * image_rate - uplift: Final = get_regional_uplift_multiplier(model_info, data_residency) + uplift: Final = get_regional_uplift_multiplier(model_info, data_residency) * get_fedramp_uplift_multiplier( + model_info, data_residency + ) if uplift != 1.0: total_prompt_cost *= uplift total_completion_cost *= uplift diff --git a/litellm/litellm_core_utils/llm_cost_calc/utils.py b/litellm/litellm_core_utils/llm_cost_calc/utils.py index d8f9a909e18..9ac4f9d5ee8 100644 --- a/litellm/litellm_core_utils/llm_cost_calc/utils.py +++ b/litellm/litellm_core_utils/llm_cost_calc/utils.py @@ -1441,6 +1441,29 @@ def get_regional_uplift_multiplier(model_info: ModelInfo, data_residency: str | _get_regional_uplift_multiplier = get_regional_uplift_multiplier +def get_fedramp_uplift_multiplier(model_info: ModelInfo, data_residency: str | None) -> float: + """ + Resolve the per-model uplift multiplier for the OpenAI FedRAMP endpoint. + + OpenAI applies a flat percentage uplift (e.g. +10%) on all token costs for + requests served from gov.api.openai.com. The multiplier is stored on the + model entry as ``fedramp_uplift_multiplier`` (e.g. 1.10). + + Returns 1.0 (no uplift) unless ``data_residency`` is ``"fedramp"`` and the + model has a multiplier configured. + """ + if data_residency is None or data_residency.lower() != DataResidency.FEDRAMP.value: + return 1.0 + multiplier: Final = model_info.get("fedramp_uplift_multiplier") + if multiplier is None: + return 1.0 + try: + return float(multiplier) + except (TypeError, ValueError): + verbose_logger.exception("Invalid fedramp_uplift_multiplier for model; defaulting to 1.0") + return 1.0 + + def get_vertex_regional_endpoint_uplift(model_info: ModelInfo, vertex_location: str | None) -> float: """ Resolve the per-model uplift multiplier for Vertex AI non-global (regional and @@ -1550,8 +1573,8 @@ def generic_cost_per_token( Input: - model: str, the model name without provider prefix - usage: LiteLLM Usage block, containing anthropic caching information - - data_residency: optional OpenAI data-residency region (e.g. "eu", "us"), - used to apply the per-model regional-processing uplift multiplier. + - data_residency: optional OpenAI data-residency region (e.g. "eu", "us") or + "fedramp", used to apply the per-model regional-processing or FedRAMP uplift multiplier. - vertex_location: optional Vertex AI location the request was served from (e.g. "us-east5", "global"), used to apply the per-model regional-endpoint uplift multiplier when non-global. @@ -1728,10 +1751,9 @@ def generic_cost_per_token( ) completion_cost += float(video_tokens) * _output_cost_per_video_token - ## REGIONAL DATA-RESIDENCY UPLIFT - # Applied as a flat multiplier across all token costs for the request - # when the upstream is a regionalized OpenAI host (eu./us.api.openai.com). - uplift: Final = get_regional_uplift_multiplier(resolved_model_info, data_residency) + uplift: Final = get_regional_uplift_multiplier(resolved_model_info, data_residency) * get_fedramp_uplift_multiplier( + resolved_model_info, data_residency + ) if uplift != 1.0: prompt_cost *= uplift completion_cost *= uplift @@ -1874,6 +1896,7 @@ def _cost_map_billed_rates( ) multiplier: Final = ( get_regional_uplift_multiplier(model_info, data_residency) + * get_fedramp_uplift_multiplier(model_info, data_residency) * get_vertex_regional_endpoint_uplift(model_info, vertex_location) * get_provider_specific_geo_multiplier(model_info=model_info, usage=usage) * get_provider_specific_speed_multiplier(model_info=model_info, usage=usage) @@ -2059,8 +2082,10 @@ def calculate_prompt_caching_savings( {ttl: rate - prompt_base_cost for ttl, rate in base_costs.cache_creation_rates_by_ttl.items()} ), ) - uplift: Final = get_regional_uplift_multiplier(model_info, data_residency) * get_vertex_regional_endpoint_uplift( - model_info, vertex_location + uplift: Final = ( + get_regional_uplift_multiplier(model_info, data_residency) + * get_fedramp_uplift_multiplier(model_info, data_residency) + * get_vertex_regional_endpoint_uplift(model_info, vertex_location) ) return (read_discount - write_premium) * uplift diff --git a/litellm/llms/openai/cost_calculation.py b/litellm/llms/openai/cost_calculation.py index 36d7e4d95de..bc8aaab3185 100644 --- a/litellm/llms/openai/cost_calculation.py +++ b/litellm/llms/openai/cost_calculation.py @@ -31,9 +31,9 @@ def cost_per_token( Input: - model: str, the model name without provider prefix - usage: LiteLLM Usage block, containing anthropic caching information - - data_residency: optional OpenAI data-residency region (e.g. "eu", "us"), - inferred from api_base. Applies the model's regional-processing - uplift multiplier when set. + - data_residency: optional OpenAI data-residency mode (e.g. "eu", "us", "fedramp"), + inferred from api_base, including gov.api.openai.com. Applies the + model's regional-processing uplift multiplier when set. Returns: Tuple[float, float] - prompt_cost_in_usd, completion_cost_in_usd diff --git a/litellm/llms/openai/data_residency.py b/litellm/llms/openai/data_residency.py index d4859221541..a8831db3517 100644 --- a/litellm/llms/openai/data_residency.py +++ b/litellm/llms/openai/data_residency.py @@ -7,9 +7,12 @@ enabled and rejects requests sent to the wrong host, so the api_base hostname is the authoritative signal of which region a request was processed in. """ +import os from typing import Final from urllib.parse import urlparse +import litellm + # Mapping of OpenAI regional hostnames to the corresponding data-residency # value used by the cost calculator. See # https://developers.openai.com/api/docs/pricing for the regional-processing @@ -17,6 +20,7 @@ from urllib.parse import urlparse _OPENAI_REGIONAL_HOSTS: Final[dict[str, str]] = { "eu.api.openai.com": "eu", "us.api.openai.com": "us", + "gov.api.openai.com": "fedramp", } @@ -25,13 +29,22 @@ def infer_openai_data_residency(custom_llm_provider: str | None, api_base: str | Derive the OpenAI data-residency region from an api_base URL. Returns ``"eu"`` for the EU regional host, ``"us"`` for the US regional - host, and ``None`` for the default global host, any non-OpenAI provider, - or any non-OpenAI URL. + host, ``"fedramp"`` for ``gov.api.openai.com``, and ``None`` for the + default global host, any non-OpenAI provider, or any non-OpenAI URL. + + When ``api_base`` is unset, falls back to the same defaults the OpenAI + handler uses (``litellm.api_base``, ``OPENAI_BASE_URL``, ``OPENAI_API_BASE``) + so env-configured regional endpoints are billed with their uplift. """ - if custom_llm_provider != "openai" or not api_base: + if custom_llm_provider != "openai": + return None + resolved_api_base: Final = ( + api_base or litellm.api_base or os.getenv("OPENAI_BASE_URL") or os.getenv("OPENAI_API_BASE") + ) + if not resolved_api_base: return None try: - host: Final = urlparse(api_base).hostname + host: Final = urlparse(resolved_api_base).hostname except (TypeError, ValueError): return None if not host: diff --git a/litellm/main.py b/litellm/main.py index d0b34614353..8cb133bf76b 100644 --- a/litellm/main.py +++ b/litellm/main.py @@ -6111,6 +6111,11 @@ def completion( messages if readable_messages is None else list(readable_messages) ) + supplied_client_api_base: Final = ( + str(client.base_url) + if custom_llm_provider == "openai" and isinstance(client, openai.OpenAI | openai.AsyncOpenAI) + else None + ) # For logging - save the values of the litellm-specific params passed in requested_litellm_params: Final = get_litellm_params( acompletion=acompletion, @@ -6119,7 +6124,7 @@ def completion( logger_fn=logger_fn, verbose=verbose, custom_llm_provider=custom_llm_provider, - api_base=api_base, + api_base=supplied_client_api_base or api_base, litellm_call_id=kwargs.get("litellm_call_id", None), model_alias_map=litellm.model_alias_map, completion_call_id=id, diff --git a/litellm/model_prices_and_context_window_backup.json b/litellm/model_prices_and_context_window_backup.json index c69f8b937da..7419f48f3fb 100644 --- a/litellm/model_prices_and_context_window_backup.json +++ b/litellm/model_prices_and_context_window_backup.json @@ -17799,6 +17799,7 @@ "max_output_tokens": 1024, "mode": "chat", "output_cost_per_token": 1.2e-05, + "fedramp_uplift_multiplier": 1.1, "supported_endpoints": [ "/v1/responses" ], @@ -17830,6 +17831,7 @@ "max_output_tokens": 1024, "mode": "chat", "output_cost_per_token": 1.2e-05, + "fedramp_uplift_multiplier": 1.1, "supported_endpoints": [ "/v1/responses" ], @@ -29263,6 +29265,7 @@ "max_output_tokens": 4096, "mode": "chat", "output_cost_per_token": 6e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 3e-06, "source": "https://developers.openai.com/api/docs/pricing", "supports_system_messages": true, @@ -29278,6 +29281,7 @@ "max_output_tokens": 4096, "mode": "chat", "output_cost_per_token": 6e-06, + "fedramp_uplift_multiplier": 1.1, "supports_system_messages": true, "supports_tool_choice": true }, @@ -29291,6 +29295,7 @@ "max_output_tokens": 4096, "mode": "chat", "output_cost_per_token": 6e-06, + "fedramp_uplift_multiplier": 1.1, "supports_system_messages": true, "supports_tool_choice": true }, @@ -29304,6 +29309,7 @@ "max_output_tokens": 4096, "mode": "chat", "output_cost_per_token": 6e-06, + "fedramp_uplift_multiplier": 1.1, "supports_system_messages": true, "supports_tool_choice": true }, @@ -29317,6 +29323,7 @@ "max_output_tokens": 4096, "mode": "chat", "output_cost_per_token": 6e-05, + "fedramp_uplift_multiplier": 1.1, "source": "OpenAI needs to add pricing for this ft model, will be updated when added by OpenAI. Defaulting to base model pricing", "supports_function_calling": true, "supports_system_messages": true, @@ -29335,6 +29342,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 1.5e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 1.25e-05, "source": "https://developers.openai.com/api/docs/pricing", "supports_function_calling": true, @@ -29357,6 +29365,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 1.5e-05, + "fedramp_uplift_multiplier": 1.1, "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_pdf_input": true, @@ -29378,6 +29387,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 1.2e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 6e-07, "source": "https://developers.openai.com/api/docs/pricing", "supports_function_calling": true, @@ -29401,6 +29411,7 @@ "max_output_tokens": 32768, "mode": "chat", "output_cost_per_token": 1.2e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 6e-06, "source": "https://developers.openai.com/api/docs/pricing", "supports_function_calling": true, @@ -29423,6 +29434,7 @@ "max_output_tokens": 32768, "mode": "chat", "output_cost_per_token": 3.2e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 1.6e-06, "source": "https://developers.openai.com/api/docs/pricing", "supports_function_calling": true, @@ -29445,6 +29457,7 @@ "max_output_tokens": 32768, "mode": "chat", "output_cost_per_token": 8e-07, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 4e-07, "source": "https://developers.openai.com/api/docs/pricing", "supports_function_calling": true, @@ -29467,6 +29480,7 @@ "max_output_tokens": 100000, "mode": "chat", "output_cost_per_token": 1.6e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 8e-06, "source": "https://developers.openai.com/api/docs/pricing", "supports_function_calling": true, @@ -35122,6 +35136,7 @@ "max_output_tokens": 4096, "mode": "chat", "output_cost_per_token": 1.5e-06, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/pricing", "supports_function_calling": true, "supports_prompt_caching": true, @@ -35139,6 +35154,7 @@ "max_output_tokens": 4096, "mode": "chat", "output_cost_per_token": 1.5e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 7.5e-07, "source": "https://developers.openai.com/api/docs/pricing", "supports_function_calling": true, @@ -35158,6 +35174,7 @@ "max_output_tokens": 4096, "mode": "chat", "output_cost_per_token": 2e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 2e-06, "source": "https://developers.openai.com/api/docs/pricing", "supports_function_calling": true, @@ -35175,6 +35192,7 @@ "max_output_tokens": 4096, "mode": "chat", "output_cost_per_token": 4e-06, + "fedramp_uplift_multiplier": 1.1, "supports_prompt_caching": true, "supports_system_messages": true, "supports_tool_choice": true @@ -35209,6 +35227,7 @@ "max_output_tokens": 4096, "mode": "chat", "output_cost_per_token": 6e-05, + "fedramp_uplift_multiplier": 1.1, "supports_function_calling": true, "supports_prompt_caching": true, "supports_system_messages": true, @@ -35225,6 +35244,7 @@ "max_output_tokens": 4096, "mode": "chat", "output_cost_per_token": 6e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 3e-05, "source": "https://developers.openai.com/api/docs/pricing", "supports_function_calling": true, @@ -35242,6 +35262,7 @@ "max_output_tokens": 4096, "mode": "chat", "output_cost_per_token": 3e-05, + "fedramp_uplift_multiplier": 1.1, "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_prompt_caching": true, @@ -35258,6 +35279,7 @@ "max_output_tokens": 4096, "mode": "chat", "output_cost_per_token": 3e-05, + "fedramp_uplift_multiplier": 1.1, "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_pdf_input": true, @@ -35277,6 +35299,7 @@ "max_output_tokens": 4096, "mode": "chat", "output_cost_per_token": 3e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 1.5e-05, "source": "https://developers.openai.com/api/docs/pricing", "supports_function_calling": true, @@ -35301,6 +35324,7 @@ "max_output_tokens": 32768, "mode": "chat", "output_cost_per_token": 8e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 4e-06, "output_cost_per_token_priority": 1.4e-05, "search_context_cost_per_query": { @@ -35346,6 +35370,7 @@ "max_output_tokens": 32768, "mode": "chat", "output_cost_per_token": 8e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_priority": 1.4e-05, "output_cost_per_token_batches": 4e-06, "search_context_cost_per_query": { @@ -35391,6 +35416,7 @@ "max_output_tokens": 32768, "mode": "chat", "output_cost_per_token": 1.6e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 8e-07, "output_cost_per_token_priority": 2.8e-06, "search_context_cost_per_query": { @@ -35436,6 +35462,7 @@ "max_output_tokens": 32768, "mode": "chat", "output_cost_per_token": 1.6e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_priority": 2.8e-06, "output_cost_per_token_batches": 8e-07, "search_context_cost_per_query": { @@ -35481,6 +35508,7 @@ "max_output_tokens": 32768, "mode": "chat", "output_cost_per_token": 4e-07, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 2e-07, "output_cost_per_token_priority": 8e-07, "source": "https://developers.openai.com/api/docs/pricing", @@ -35520,6 +35548,7 @@ "max_output_tokens": 32768, "mode": "chat", "output_cost_per_token": 4e-07, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_priority": 8e-07, "output_cost_per_token_batches": 2e-07, "source": "https://developers.openai.com/api/docs/pricing", @@ -35559,6 +35588,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 5e-06, "output_cost_per_token_priority": 1.7e-05, "source": "https://developers.openai.com/api/docs/pricing", @@ -35583,6 +35613,7 @@ "max_output_tokens": 4096, "mode": "chat", "output_cost_per_token": 1.5e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 7.5e-06, "output_cost_per_token_priority": 2.625e-05, "source": "https://developers.openai.com/api/docs/pricing", @@ -35608,6 +35639,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_priority": 1.7e-05, "output_cost_per_token_batches": 5e-06, "source": "https://developers.openai.com/api/docs/pricing", @@ -35634,6 +35666,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_priority": 1.7e-05, "output_cost_per_token_batches": 5e-06, "source": "https://developers.openai.com/api/docs/pricing", @@ -35926,6 +35959,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 6e-07, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 3e-07, "output_cost_per_token_priority": 1e-06, "supports_function_calling": true, @@ -35957,6 +35991,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 6e-07, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_priority": 1e-06, "output_cost_per_token_batches": 3e-07, "search_context_cost_per_query": { @@ -36021,6 +36056,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 6e-07, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 3e-07, "search_context_cost_per_query": { "search_context_size_high": 0.025, @@ -36049,6 +36085,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 6e-07, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 3e-07, "search_context_cost_per_query": { "search_context_size_high": 0.025, @@ -36136,6 +36173,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 5e-06, "search_context_cost_per_query": { "search_context_size_high": 0.025, @@ -36164,6 +36202,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 5e-06, "search_context_cost_per_query": { "search_context_size_high": 0.025, @@ -36786,6 +36825,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 5e-06, "output_cost_per_token_flex": 5e-06, "output_cost_per_token_priority": 2e-05, @@ -36840,6 +36880,7 @@ "max_output_tokens": 128000, "mode": "responses", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/models/gpt-5-codex", "supported_endpoints": [ "/v1/responses" @@ -36874,6 +36915,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_priority": 2e-05, "reasoning_contexts": [ "auto", @@ -36934,6 +36976,7 @@ "max_output_tokens": 128000, "mode": "responses", "output_cost_per_token": 2e-06, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/models/gpt-5.1-codex-mini", "supported_endpoints": [ "/v1/responses" @@ -36966,6 +37009,7 @@ "max_output_tokens": 128000, "mode": "responses", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/models/gpt-5.1-codex-max", "supported_endpoints": [ "/v1/responses" @@ -36999,6 +37043,7 @@ "max_output_tokens": 128000, "mode": "responses", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/models/gpt-5.1-codex", "supported_endpoints": [ "/v1/responses" @@ -37032,6 +37077,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/models/gpt-5.1-chat-latest", "supported_endpoints": [ "/v1/chat/completions", @@ -37069,6 +37115,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_priority": 2e-05, "reasoning_contexts": [ "auto", @@ -37127,6 +37174,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 1.4e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_priority": 2.8e-05, "reasoning_contexts": [ "auto", @@ -37186,6 +37234,7 @@ "max_output_tokens": 128000, "mode": "responses", "output_cost_per_token": 1.4e-05, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/models/gpt-5.2-codex", "supported_endpoints": [ "/v1/responses" @@ -37219,6 +37268,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 1.4e-05, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/models/gpt-5.2-chat-latest", "supported_endpoints": [ "/v1/chat/completions", @@ -37253,6 +37303,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 1.4e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_priority": 2.8e-05, "reasoning_contexts": [ "auto", @@ -37309,6 +37360,7 @@ "max_output_tokens": 128000, "mode": "responses", "output_cost_per_token": 0.000168, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 8.4e-05, "reasoning_contexts": [ "auto", @@ -37357,6 +37409,7 @@ "max_output_tokens": 128000, "mode": "responses", "output_cost_per_token": 0.000168, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 8.4e-05, "reasoning_contexts": [ "auto", @@ -37463,6 +37516,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -37556,6 +37610,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -37657,6 +37712,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -37749,6 +37805,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -37839,6 +37896,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -37931,6 +37989,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -38022,6 +38081,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -38070,6 +38130,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 7.5e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_above_272k_tokens": 0.0001125, "supported_endpoints": [ "/v1/chat/completions", @@ -38108,6 +38169,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 7.5e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_above_272k_tokens": 0.0001125, "supported_endpoints": [ "/v1/chat/completions", @@ -38146,6 +38208,7 @@ "max_output_tokens": 128000, "mode": "responses", "output_cost_per_token": 7.5e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_above_272k_tokens": 0.0001125, "supported_endpoints": [ "/v1/responses" @@ -38183,6 +38246,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 2e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_above_272k_tokens": 3e-05, "supported_endpoints": [ "/v1/chat/completions", @@ -38221,6 +38285,7 @@ "max_output_tokens": 128000, "mode": "responses", "output_cost_per_token": 2e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_above_272k_tokens": 3e-05, "supported_endpoints": [ "/v1/responses" @@ -38254,6 +38319,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 3e-05, + "fedramp_uplift_multiplier": 1.1, "reasoning_effort_levels": [ "medium" ], @@ -38322,6 +38388,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -38392,6 +38459,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -38454,6 +38522,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -38513,6 +38582,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -38580,6 +38650,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "supported_endpoints": [ "/v1/chat/completions", "/v1/batch", @@ -38645,6 +38716,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "supported_endpoints": [ "/v1/chat/completions", "/v1/batch", @@ -38702,6 +38774,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -38762,6 +38835,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -38824,6 +38898,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -38886,6 +38961,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -38945,6 +39021,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -39009,6 +39086,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -39054,6 +39132,7 @@ "max_output_tokens": 272000, "mode": "responses", "output_cost_per_token": 0.00012, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 6e-05, "reasoning_contexts": [ "auto", @@ -39105,6 +39184,7 @@ "max_output_tokens": 272000, "mode": "responses", "output_cost_per_token": 0.00012, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 6e-05, "reasoning_contexts": [ "auto", @@ -39162,6 +39242,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 5e-06, "output_cost_per_token_flex": 5e-06, "output_cost_per_token_priority": 2e-05, @@ -39216,6 +39297,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "supported_endpoints": [ "/v1/chat/completions", "/v1/batch", @@ -39253,6 +39335,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/models/gpt-5-chat-latest", "supported_endpoints": [ "/v1/chat/completions", @@ -39285,6 +39368,7 @@ "max_output_tokens": 128000, "mode": "responses", "output_cost_per_token": 1.4e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_priority": 2.8e-05, "reasoning_contexts": [ "auto", @@ -39339,6 +39423,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 1.4e-05, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/models/gpt-5.3-chat-latest", "supported_endpoints": [ "/v1/chat/completions", @@ -39378,6 +39463,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 2e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 1e-06, "output_cost_per_token_flex": 1e-06, "output_cost_per_token_priority": 3.6e-06, @@ -39438,6 +39524,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 2e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 1e-06, "output_cost_per_token_flex": 1e-06, "output_cost_per_token_priority": 3.6e-06, @@ -39496,6 +39583,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 4e-07, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 2e-07, "output_cost_per_token_flex": 2e-07, "reasoning_contexts": [ @@ -39553,6 +39641,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 4e-07, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 2e-07, "output_cost_per_token_flex": 2e-07, "reasoning_contexts": [ @@ -45472,6 +45561,7 @@ "max_output_tokens": 100000, "mode": "chat", "output_cost_per_token": 6e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 3e-05, "source": "https://developers.openai.com/api/docs/pricing", "supports_function_calling": true, @@ -45496,6 +45586,7 @@ "max_output_tokens": 100000, "mode": "chat", "output_cost_per_token": 6e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 3e-05, "source": "https://developers.openai.com/api/docs/pricing", "supports_function_calling": true, @@ -45519,6 +45610,7 @@ "max_output_tokens": 100000, "mode": "responses", "output_cost_per_token": 0.0006, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 0.0003, "source": "https://developers.openai.com/api/docs/pricing", "supported_endpoints": [ @@ -45554,6 +45646,7 @@ "max_output_tokens": 100000, "mode": "responses", "output_cost_per_token": 0.0006, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 0.0003, "source": "https://developers.openai.com/api/docs/pricing", "supported_endpoints": [ @@ -45593,6 +45686,7 @@ "max_output_tokens": 100000, "mode": "chat", "output_cost_per_token": 8e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 4e-06, "output_cost_per_token_flex": 4e-06, "output_cost_per_token_priority": 1.4e-05, @@ -45640,6 +45734,7 @@ "max_output_tokens": 100000, "mode": "responses", "output_cost_per_token": 4e-05, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/models/o3-deep-research", "supported_endpoints": [ "/v1/batch", @@ -45676,6 +45771,7 @@ "max_output_tokens": 100000, "mode": "responses", "output_cost_per_token": 4e-05, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/models/o3-deep-research", "supported_endpoints": [ "/v1/batch", @@ -45714,6 +45810,7 @@ "max_output_tokens": 100000, "mode": "chat", "output_cost_per_token": 8e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 4e-06, "output_cost_per_token_flex": 4e-06, "output_cost_per_token_priority": 1.4e-05, @@ -45765,6 +45862,7 @@ "max_output_tokens": 100000, "mode": "chat", "output_cost_per_token": 4.4e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 2.2e-06, "source": "https://developers.openai.com/api/docs/pricing", "supports_function_calling": true, @@ -45787,6 +45885,7 @@ "max_output_tokens": 100000, "mode": "chat", "output_cost_per_token": 4.4e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 2.2e-06, "source": "https://developers.openai.com/api/docs/pricing", "supports_function_calling": true, @@ -45808,6 +45907,7 @@ "max_output_tokens": 100000, "mode": "responses", "output_cost_per_token": 8e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 4e-05, "reasoning_contexts": [ "auto", @@ -45854,6 +45954,7 @@ "max_output_tokens": 100000, "mode": "responses", "output_cost_per_token": 8e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 4e-05, "reasoning_contexts": [ "auto", @@ -45905,6 +46006,7 @@ "max_output_tokens": 100000, "mode": "chat", "output_cost_per_token": 4.4e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 2.2e-06, "output_cost_per_token_flex": 2.2e-06, "output_cost_per_token_priority": 8e-06, @@ -45932,6 +46034,7 @@ "max_output_tokens": 100000, "mode": "responses", "output_cost_per_token": 8e-06, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/models/o4-mini-deep-research", "supported_endpoints": [ "/v1/batch", @@ -45968,6 +46071,7 @@ "max_output_tokens": 100000, "mode": "responses", "output_cost_per_token": 8e-06, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/models/o4-mini-deep-research", "supported_endpoints": [ "/v1/batch", @@ -46006,6 +46110,7 @@ "max_output_tokens": 100000, "mode": "chat", "output_cost_per_token": 4.4e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 2.2e-06, "output_cost_per_token_flex": 2.2e-06, "output_cost_per_token_priority": 8e-06, @@ -59584,6 +59689,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -59611,6 +59717,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -70793,6 +70900,7 @@ "litellm_provider": "openai", "mode": "chat", "output_cost_per_token": 7.5e-05, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/pricing", "supports_reasoning": true }, @@ -70802,6 +70910,7 @@ "litellm_provider": "openai", "mode": "chat", "output_cost_per_token": 7.5e-05, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/pricing", "supports_reasoning": true }, @@ -76155,6 +76264,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -79685,6 +79795,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": {"search_context_size_high":0.01,"search_context_size_low":0.01,"search_context_size_medium":0.01}, "source": "https://developers.openai.com/api/docs/pricing", "supported_endpoints": [ @@ -82379,6 +82490,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": {"search_context_size_high":0.01,"search_context_size_low":0.01,"search_context_size_medium":0.01}, "source": "https://developers.openai.com/api/docs/pricing", "supported_endpoints": [ diff --git a/litellm/types/utils.py b/litellm/types/utils.py index ee80b63f93b..dc71ba149da 100644 --- a/litellm/types/utils.py +++ b/litellm/types/utils.py @@ -431,6 +431,9 @@ class ModelInfoBase(ProviderSpecificModelInfo, total=False): regional_processing_uplift_multiplier_us: ( float | None ) # OpenAI US data-residency uplift multiplier applied to all token costs (e.g. 1.10 = +10%) + fedramp_uplift_multiplier: ReadOnly[ + float | None + ] # OpenAI FedRAMP endpoint (gov.api.openai.com) uplift multiplier applied to all token costs (e.g. 1.10 = +10%) regional_endpoint_uplift_multiplier: ReadOnly[ float | None ] # Vertex AI non-global (regional) endpoint uplift multiplier applied to all token costs (e.g. 1.10 = +10%) @@ -4258,6 +4261,7 @@ class CustomPricingLiteLLMParams(MirroredPricingParams): annotation_cost_per_page_batches: float | None = None regional_processing_uplift_multiplier_eu: float | None = None regional_processing_uplift_multiplier_us: float | None = None + fedramp_uplift_multiplier: float | None = None regional_endpoint_uplift_multiplier: float | None = None @classmethod @@ -5260,11 +5264,13 @@ class DataResidency(Enum): Inferred from the OpenAI api_base host (eu.api.openai.com -> EU, us.api.openai.com -> US). Used to apply the regional-processing cost uplift (see ``regional_processing_uplift_multiplier_`` - on ModelInfo). + on ModelInfo). gov.api.openai.com maps to FEDRAMP, which applies + ``fedramp_uplift_multiplier`` instead. """ US = "us" EU = "eu" + FEDRAMP = "fedramp" LLMResponseTypes = ( diff --git a/litellm/utils.py b/litellm/utils.py index ec7bb041fb5..b457da058ad 100644 --- a/litellm/utils.py +++ b/litellm/utils.py @@ -7504,6 +7504,7 @@ def get_model_info_helper( regional_processing_uplift_multiplier_us=_model_info.get( "regional_processing_uplift_multiplier_us", None ), + fedramp_uplift_multiplier=_model_info.get("fedramp_uplift_multiplier", None), regional_endpoint_uplift_multiplier=_model_info.get("regional_endpoint_uplift_multiplier", None), output_cost_per_audio_token=_model_info.get("output_cost_per_audio_token", None), output_cost_per_character=_model_info.get("output_cost_per_character", None), diff --git a/model_prices_and_context_window.json b/model_prices_and_context_window.json index c69f8b937da..7419f48f3fb 100644 --- a/model_prices_and_context_window.json +++ b/model_prices_and_context_window.json @@ -17799,6 +17799,7 @@ "max_output_tokens": 1024, "mode": "chat", "output_cost_per_token": 1.2e-05, + "fedramp_uplift_multiplier": 1.1, "supported_endpoints": [ "/v1/responses" ], @@ -17830,6 +17831,7 @@ "max_output_tokens": 1024, "mode": "chat", "output_cost_per_token": 1.2e-05, + "fedramp_uplift_multiplier": 1.1, "supported_endpoints": [ "/v1/responses" ], @@ -29263,6 +29265,7 @@ "max_output_tokens": 4096, "mode": "chat", "output_cost_per_token": 6e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 3e-06, "source": "https://developers.openai.com/api/docs/pricing", "supports_system_messages": true, @@ -29278,6 +29281,7 @@ "max_output_tokens": 4096, "mode": "chat", "output_cost_per_token": 6e-06, + "fedramp_uplift_multiplier": 1.1, "supports_system_messages": true, "supports_tool_choice": true }, @@ -29291,6 +29295,7 @@ "max_output_tokens": 4096, "mode": "chat", "output_cost_per_token": 6e-06, + "fedramp_uplift_multiplier": 1.1, "supports_system_messages": true, "supports_tool_choice": true }, @@ -29304,6 +29309,7 @@ "max_output_tokens": 4096, "mode": "chat", "output_cost_per_token": 6e-06, + "fedramp_uplift_multiplier": 1.1, "supports_system_messages": true, "supports_tool_choice": true }, @@ -29317,6 +29323,7 @@ "max_output_tokens": 4096, "mode": "chat", "output_cost_per_token": 6e-05, + "fedramp_uplift_multiplier": 1.1, "source": "OpenAI needs to add pricing for this ft model, will be updated when added by OpenAI. Defaulting to base model pricing", "supports_function_calling": true, "supports_system_messages": true, @@ -29335,6 +29342,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 1.5e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 1.25e-05, "source": "https://developers.openai.com/api/docs/pricing", "supports_function_calling": true, @@ -29357,6 +29365,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 1.5e-05, + "fedramp_uplift_multiplier": 1.1, "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_pdf_input": true, @@ -29378,6 +29387,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 1.2e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 6e-07, "source": "https://developers.openai.com/api/docs/pricing", "supports_function_calling": true, @@ -29401,6 +29411,7 @@ "max_output_tokens": 32768, "mode": "chat", "output_cost_per_token": 1.2e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 6e-06, "source": "https://developers.openai.com/api/docs/pricing", "supports_function_calling": true, @@ -29423,6 +29434,7 @@ "max_output_tokens": 32768, "mode": "chat", "output_cost_per_token": 3.2e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 1.6e-06, "source": "https://developers.openai.com/api/docs/pricing", "supports_function_calling": true, @@ -29445,6 +29457,7 @@ "max_output_tokens": 32768, "mode": "chat", "output_cost_per_token": 8e-07, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 4e-07, "source": "https://developers.openai.com/api/docs/pricing", "supports_function_calling": true, @@ -29467,6 +29480,7 @@ "max_output_tokens": 100000, "mode": "chat", "output_cost_per_token": 1.6e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 8e-06, "source": "https://developers.openai.com/api/docs/pricing", "supports_function_calling": true, @@ -35122,6 +35136,7 @@ "max_output_tokens": 4096, "mode": "chat", "output_cost_per_token": 1.5e-06, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/pricing", "supports_function_calling": true, "supports_prompt_caching": true, @@ -35139,6 +35154,7 @@ "max_output_tokens": 4096, "mode": "chat", "output_cost_per_token": 1.5e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 7.5e-07, "source": "https://developers.openai.com/api/docs/pricing", "supports_function_calling": true, @@ -35158,6 +35174,7 @@ "max_output_tokens": 4096, "mode": "chat", "output_cost_per_token": 2e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 2e-06, "source": "https://developers.openai.com/api/docs/pricing", "supports_function_calling": true, @@ -35175,6 +35192,7 @@ "max_output_tokens": 4096, "mode": "chat", "output_cost_per_token": 4e-06, + "fedramp_uplift_multiplier": 1.1, "supports_prompt_caching": true, "supports_system_messages": true, "supports_tool_choice": true @@ -35209,6 +35227,7 @@ "max_output_tokens": 4096, "mode": "chat", "output_cost_per_token": 6e-05, + "fedramp_uplift_multiplier": 1.1, "supports_function_calling": true, "supports_prompt_caching": true, "supports_system_messages": true, @@ -35225,6 +35244,7 @@ "max_output_tokens": 4096, "mode": "chat", "output_cost_per_token": 6e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 3e-05, "source": "https://developers.openai.com/api/docs/pricing", "supports_function_calling": true, @@ -35242,6 +35262,7 @@ "max_output_tokens": 4096, "mode": "chat", "output_cost_per_token": 3e-05, + "fedramp_uplift_multiplier": 1.1, "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_prompt_caching": true, @@ -35258,6 +35279,7 @@ "max_output_tokens": 4096, "mode": "chat", "output_cost_per_token": 3e-05, + "fedramp_uplift_multiplier": 1.1, "supports_function_calling": true, "supports_parallel_function_calling": true, "supports_pdf_input": true, @@ -35277,6 +35299,7 @@ "max_output_tokens": 4096, "mode": "chat", "output_cost_per_token": 3e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 1.5e-05, "source": "https://developers.openai.com/api/docs/pricing", "supports_function_calling": true, @@ -35301,6 +35324,7 @@ "max_output_tokens": 32768, "mode": "chat", "output_cost_per_token": 8e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 4e-06, "output_cost_per_token_priority": 1.4e-05, "search_context_cost_per_query": { @@ -35346,6 +35370,7 @@ "max_output_tokens": 32768, "mode": "chat", "output_cost_per_token": 8e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_priority": 1.4e-05, "output_cost_per_token_batches": 4e-06, "search_context_cost_per_query": { @@ -35391,6 +35416,7 @@ "max_output_tokens": 32768, "mode": "chat", "output_cost_per_token": 1.6e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 8e-07, "output_cost_per_token_priority": 2.8e-06, "search_context_cost_per_query": { @@ -35436,6 +35462,7 @@ "max_output_tokens": 32768, "mode": "chat", "output_cost_per_token": 1.6e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_priority": 2.8e-06, "output_cost_per_token_batches": 8e-07, "search_context_cost_per_query": { @@ -35481,6 +35508,7 @@ "max_output_tokens": 32768, "mode": "chat", "output_cost_per_token": 4e-07, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 2e-07, "output_cost_per_token_priority": 8e-07, "source": "https://developers.openai.com/api/docs/pricing", @@ -35520,6 +35548,7 @@ "max_output_tokens": 32768, "mode": "chat", "output_cost_per_token": 4e-07, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_priority": 8e-07, "output_cost_per_token_batches": 2e-07, "source": "https://developers.openai.com/api/docs/pricing", @@ -35559,6 +35588,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 5e-06, "output_cost_per_token_priority": 1.7e-05, "source": "https://developers.openai.com/api/docs/pricing", @@ -35583,6 +35613,7 @@ "max_output_tokens": 4096, "mode": "chat", "output_cost_per_token": 1.5e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 7.5e-06, "output_cost_per_token_priority": 2.625e-05, "source": "https://developers.openai.com/api/docs/pricing", @@ -35608,6 +35639,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_priority": 1.7e-05, "output_cost_per_token_batches": 5e-06, "source": "https://developers.openai.com/api/docs/pricing", @@ -35634,6 +35666,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_priority": 1.7e-05, "output_cost_per_token_batches": 5e-06, "source": "https://developers.openai.com/api/docs/pricing", @@ -35926,6 +35959,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 6e-07, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 3e-07, "output_cost_per_token_priority": 1e-06, "supports_function_calling": true, @@ -35957,6 +35991,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 6e-07, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_priority": 1e-06, "output_cost_per_token_batches": 3e-07, "search_context_cost_per_query": { @@ -36021,6 +36056,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 6e-07, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 3e-07, "search_context_cost_per_query": { "search_context_size_high": 0.025, @@ -36049,6 +36085,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 6e-07, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 3e-07, "search_context_cost_per_query": { "search_context_size_high": 0.025, @@ -36136,6 +36173,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 5e-06, "search_context_cost_per_query": { "search_context_size_high": 0.025, @@ -36164,6 +36202,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 5e-06, "search_context_cost_per_query": { "search_context_size_high": 0.025, @@ -36786,6 +36825,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 5e-06, "output_cost_per_token_flex": 5e-06, "output_cost_per_token_priority": 2e-05, @@ -36840,6 +36880,7 @@ "max_output_tokens": 128000, "mode": "responses", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/models/gpt-5-codex", "supported_endpoints": [ "/v1/responses" @@ -36874,6 +36915,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_priority": 2e-05, "reasoning_contexts": [ "auto", @@ -36934,6 +36976,7 @@ "max_output_tokens": 128000, "mode": "responses", "output_cost_per_token": 2e-06, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/models/gpt-5.1-codex-mini", "supported_endpoints": [ "/v1/responses" @@ -36966,6 +37009,7 @@ "max_output_tokens": 128000, "mode": "responses", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/models/gpt-5.1-codex-max", "supported_endpoints": [ "/v1/responses" @@ -36999,6 +37043,7 @@ "max_output_tokens": 128000, "mode": "responses", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/models/gpt-5.1-codex", "supported_endpoints": [ "/v1/responses" @@ -37032,6 +37077,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/models/gpt-5.1-chat-latest", "supported_endpoints": [ "/v1/chat/completions", @@ -37069,6 +37115,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_priority": 2e-05, "reasoning_contexts": [ "auto", @@ -37127,6 +37174,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 1.4e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_priority": 2.8e-05, "reasoning_contexts": [ "auto", @@ -37186,6 +37234,7 @@ "max_output_tokens": 128000, "mode": "responses", "output_cost_per_token": 1.4e-05, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/models/gpt-5.2-codex", "supported_endpoints": [ "/v1/responses" @@ -37219,6 +37268,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 1.4e-05, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/models/gpt-5.2-chat-latest", "supported_endpoints": [ "/v1/chat/completions", @@ -37253,6 +37303,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 1.4e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_priority": 2.8e-05, "reasoning_contexts": [ "auto", @@ -37309,6 +37360,7 @@ "max_output_tokens": 128000, "mode": "responses", "output_cost_per_token": 0.000168, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 8.4e-05, "reasoning_contexts": [ "auto", @@ -37357,6 +37409,7 @@ "max_output_tokens": 128000, "mode": "responses", "output_cost_per_token": 0.000168, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 8.4e-05, "reasoning_contexts": [ "auto", @@ -37463,6 +37516,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -37556,6 +37610,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -37657,6 +37712,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -37749,6 +37805,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -37839,6 +37896,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -37931,6 +37989,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -38022,6 +38081,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -38070,6 +38130,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 7.5e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_above_272k_tokens": 0.0001125, "supported_endpoints": [ "/v1/chat/completions", @@ -38108,6 +38169,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 7.5e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_above_272k_tokens": 0.0001125, "supported_endpoints": [ "/v1/chat/completions", @@ -38146,6 +38208,7 @@ "max_output_tokens": 128000, "mode": "responses", "output_cost_per_token": 7.5e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_above_272k_tokens": 0.0001125, "supported_endpoints": [ "/v1/responses" @@ -38183,6 +38246,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 2e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_above_272k_tokens": 3e-05, "supported_endpoints": [ "/v1/chat/completions", @@ -38221,6 +38285,7 @@ "max_output_tokens": 128000, "mode": "responses", "output_cost_per_token": 2e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_above_272k_tokens": 3e-05, "supported_endpoints": [ "/v1/responses" @@ -38254,6 +38319,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 3e-05, + "fedramp_uplift_multiplier": 1.1, "reasoning_effort_levels": [ "medium" ], @@ -38322,6 +38388,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -38392,6 +38459,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -38454,6 +38522,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -38513,6 +38582,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -38580,6 +38650,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "supported_endpoints": [ "/v1/chat/completions", "/v1/batch", @@ -38645,6 +38716,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "supported_endpoints": [ "/v1/chat/completions", "/v1/batch", @@ -38702,6 +38774,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -38762,6 +38835,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -38824,6 +38898,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -38886,6 +38961,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -38945,6 +39021,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -39009,6 +39086,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -39054,6 +39132,7 @@ "max_output_tokens": 272000, "mode": "responses", "output_cost_per_token": 0.00012, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 6e-05, "reasoning_contexts": [ "auto", @@ -39105,6 +39184,7 @@ "max_output_tokens": 272000, "mode": "responses", "output_cost_per_token": 0.00012, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 6e-05, "reasoning_contexts": [ "auto", @@ -39162,6 +39242,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 5e-06, "output_cost_per_token_flex": 5e-06, "output_cost_per_token_priority": 2e-05, @@ -39216,6 +39297,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "supported_endpoints": [ "/v1/chat/completions", "/v1/batch", @@ -39253,6 +39335,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/models/gpt-5-chat-latest", "supported_endpoints": [ "/v1/chat/completions", @@ -39285,6 +39368,7 @@ "max_output_tokens": 128000, "mode": "responses", "output_cost_per_token": 1.4e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_priority": 2.8e-05, "reasoning_contexts": [ "auto", @@ -39339,6 +39423,7 @@ "max_output_tokens": 16384, "mode": "chat", "output_cost_per_token": 1.4e-05, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/models/gpt-5.3-chat-latest", "supported_endpoints": [ "/v1/chat/completions", @@ -39378,6 +39463,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 2e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 1e-06, "output_cost_per_token_flex": 1e-06, "output_cost_per_token_priority": 3.6e-06, @@ -39438,6 +39524,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 2e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 1e-06, "output_cost_per_token_flex": 1e-06, "output_cost_per_token_priority": 3.6e-06, @@ -39496,6 +39583,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 4e-07, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 2e-07, "output_cost_per_token_flex": 2e-07, "reasoning_contexts": [ @@ -39553,6 +39641,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 4e-07, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 2e-07, "output_cost_per_token_flex": 2e-07, "reasoning_contexts": [ @@ -45472,6 +45561,7 @@ "max_output_tokens": 100000, "mode": "chat", "output_cost_per_token": 6e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 3e-05, "source": "https://developers.openai.com/api/docs/pricing", "supports_function_calling": true, @@ -45496,6 +45586,7 @@ "max_output_tokens": 100000, "mode": "chat", "output_cost_per_token": 6e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 3e-05, "source": "https://developers.openai.com/api/docs/pricing", "supports_function_calling": true, @@ -45519,6 +45610,7 @@ "max_output_tokens": 100000, "mode": "responses", "output_cost_per_token": 0.0006, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 0.0003, "source": "https://developers.openai.com/api/docs/pricing", "supported_endpoints": [ @@ -45554,6 +45646,7 @@ "max_output_tokens": 100000, "mode": "responses", "output_cost_per_token": 0.0006, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 0.0003, "source": "https://developers.openai.com/api/docs/pricing", "supported_endpoints": [ @@ -45593,6 +45686,7 @@ "max_output_tokens": 100000, "mode": "chat", "output_cost_per_token": 8e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 4e-06, "output_cost_per_token_flex": 4e-06, "output_cost_per_token_priority": 1.4e-05, @@ -45640,6 +45734,7 @@ "max_output_tokens": 100000, "mode": "responses", "output_cost_per_token": 4e-05, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/models/o3-deep-research", "supported_endpoints": [ "/v1/batch", @@ -45676,6 +45771,7 @@ "max_output_tokens": 100000, "mode": "responses", "output_cost_per_token": 4e-05, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/models/o3-deep-research", "supported_endpoints": [ "/v1/batch", @@ -45714,6 +45810,7 @@ "max_output_tokens": 100000, "mode": "chat", "output_cost_per_token": 8e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 4e-06, "output_cost_per_token_flex": 4e-06, "output_cost_per_token_priority": 1.4e-05, @@ -45765,6 +45862,7 @@ "max_output_tokens": 100000, "mode": "chat", "output_cost_per_token": 4.4e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 2.2e-06, "source": "https://developers.openai.com/api/docs/pricing", "supports_function_calling": true, @@ -45787,6 +45885,7 @@ "max_output_tokens": 100000, "mode": "chat", "output_cost_per_token": 4.4e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 2.2e-06, "source": "https://developers.openai.com/api/docs/pricing", "supports_function_calling": true, @@ -45808,6 +45907,7 @@ "max_output_tokens": 100000, "mode": "responses", "output_cost_per_token": 8e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 4e-05, "reasoning_contexts": [ "auto", @@ -45854,6 +45954,7 @@ "max_output_tokens": 100000, "mode": "responses", "output_cost_per_token": 8e-05, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 4e-05, "reasoning_contexts": [ "auto", @@ -45905,6 +46006,7 @@ "max_output_tokens": 100000, "mode": "chat", "output_cost_per_token": 4.4e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 2.2e-06, "output_cost_per_token_flex": 2.2e-06, "output_cost_per_token_priority": 8e-06, @@ -45932,6 +46034,7 @@ "max_output_tokens": 100000, "mode": "responses", "output_cost_per_token": 8e-06, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/models/o4-mini-deep-research", "supported_endpoints": [ "/v1/batch", @@ -45968,6 +46071,7 @@ "max_output_tokens": 100000, "mode": "responses", "output_cost_per_token": 8e-06, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/models/o4-mini-deep-research", "supported_endpoints": [ "/v1/batch", @@ -46006,6 +46110,7 @@ "max_output_tokens": 100000, "mode": "chat", "output_cost_per_token": 4.4e-06, + "fedramp_uplift_multiplier": 1.1, "output_cost_per_token_batches": 2.2e-06, "output_cost_per_token_flex": 2.2e-06, "output_cost_per_token_priority": 8e-06, @@ -59584,6 +59689,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -59611,6 +59717,7 @@ "max_output_tokens": 128000, "mode": "chat", "output_cost_per_token": 1e-05, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -70793,6 +70900,7 @@ "litellm_provider": "openai", "mode": "chat", "output_cost_per_token": 7.5e-05, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/pricing", "supports_reasoning": true }, @@ -70802,6 +70910,7 @@ "litellm_provider": "openai", "mode": "chat", "output_cost_per_token": 7.5e-05, + "fedramp_uplift_multiplier": 1.1, "source": "https://developers.openai.com/api/docs/pricing", "supports_reasoning": true }, @@ -76155,6 +76264,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": { "search_context_size_high": 0.01, "search_context_size_low": 0.01, @@ -79685,6 +79795,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": {"search_context_size_high":0.01,"search_context_size_low":0.01,"search_context_size_medium":0.01}, "source": "https://developers.openai.com/api/docs/pricing", "supported_endpoints": [ @@ -82379,6 +82490,7 @@ ], "regional_processing_uplift_multiplier_eu": 1.1, "regional_processing_uplift_multiplier_us": 1.1, + "fedramp_uplift_multiplier": 1.1, "search_context_cost_per_query": {"search_context_size_high":0.01,"search_context_size_low":0.01,"search_context_size_medium":0.01}, "source": "https://developers.openai.com/api/docs/pricing", "supported_endpoints": [ diff --git a/tests/unit/batches/test_batch_utils.py b/tests/unit/batches/test_batch_utils.py index b32afd51fd3..16b8cf3f0e1 100644 --- a/tests/unit/batches/test_batch_utils.py +++ b/tests/unit/batches/test_batch_utils.py @@ -2766,9 +2766,9 @@ def test_aggregate_batch_cost_uses_custom_model_info(): assert result.cost == pytest.approx(expected), f"Expected total cost {expected}, got {result.cost}" @pytest.mark.usefixtures("_vcr_outcome_gate", "setup_and_teardown") -@pytest.mark.parametrize("data_residency", ["eu", "us"]) +@pytest.mark.parametrize("data_residency", ["eu", "us", "fedramp"]) def test_batch_cost_calculator_applies_data_residency_uplift(data_residency, monkeypatch): - """batch_cost_calculator should apply the regional uplift multiplier when + """batch_cost_calculator should apply the regional or FedRAMP uplift multiplier when data_residency is set and the model carries a configured multiplier.""" monkeypatch.setenv("LITELLM_LOCAL_MODEL_COST_MAP", "True") prev_model_cost = litellm.model_cost diff --git a/tests/unit/litellm_core_utils/llm_cost_calc/test_llm_cost_calc_utils.py b/tests/unit/litellm_core_utils/llm_cost_calc/test_llm_cost_calc_utils.py index 0a08142c66f..609b28270d7 100644 --- a/tests/unit/litellm_core_utils/llm_cost_calc/test_llm_cost_calc_utils.py +++ b/tests/unit/litellm_core_utils/llm_cost_calc/test_llm_cost_calc_utils.py @@ -33,6 +33,7 @@ from litellm.types.llms.base import CachedTokensDetails from litellm.types.utils import ( CacheCreationTokenDetails, CompletionTokensDetailsWrapper, + DataResidency, ImageObject, ImageResponse, ImageUsage, @@ -2579,6 +2580,51 @@ def test_data_residency_applies_uplift(data_residency, model, _local_model_cost_ assert regional[1] == pytest.approx(base[1] * 1.10, rel=1e-9) +def test_fedramp_data_residency_applies_configured_uplift() -> None: + usage: Final = Usage(prompt_tokens=1000, completion_tokens=500, total_tokens=1500) + model_info: Final[ModelInfo] = { + "key": "fedramp-fixture", + "max_tokens": None, + "max_input_tokens": None, + "max_output_tokens": None, + "input_cost_per_token": 1e-6, + "output_cost_per_token": 2e-6, + "litellm_provider": "openai", + "mode": "chat", + "supported_openai_params": None, + "regional_processing_uplift_multiplier_us": 1.5, + "fedramp_uplift_multiplier": 1.1, + } + + base: Final = generic_cost_per_token( + model="fedramp-fixture", + usage=usage, + custom_llm_provider="openai", + model_info=model_info, + ) + fedramp: Final = generic_cost_per_token( + model="fedramp-fixture", + usage=usage, + custom_llm_provider="openai", + data_residency=DataResidency.FEDRAMP.value, + model_info=model_info, + ) + + us: Final = generic_cost_per_token( + model="fedramp-fixture", + usage=usage, + custom_llm_provider="openai", + data_residency=DataResidency.US.value, + model_info=model_info, + ) + + assert sum(base) > 0 + assert fedramp[0] == pytest.approx(base[0] * 1.1) + assert fedramp[1] == pytest.approx(base[1] * 1.1) + assert us[0] == pytest.approx(base[0] * 1.5) + assert us[1] == pytest.approx(base[1] * 1.5) + + @pytest.mark.parametrize("model", ["gpt-5", "gpt-5-mini", "gpt-5-nano", "gpt-5-pro", "gpt-4o", "gpt-4.1"]) def test_data_residency_no_uplift_for_pre_march_2026_models(model, _local_model_cost_map): """Models released before 2026-03-05 must not have the regional uplift.""" @@ -2719,6 +2765,22 @@ def test_vertex_uplift_invalid_multiplier_defaults_to_one(): ) +@pytest.mark.parametrize( + "model_info, data_residency", + [ + ({"fedramp_uplift_multiplier": "not-a-number"}, "fedramp"), + ({}, "fedramp"), + ({"fedramp_uplift_multiplier": 1.1}, "us"), + ({"fedramp_uplift_multiplier": 1.1}, None), + ], +) +def test_fedramp_uplift_defaults_to_one(model_info: ModelInfo, data_residency: str | None): + """A malformed or missing FedRAMP multiplier, or a non-FedRAMP host, bills base pricing and never raises.""" + from litellm.litellm_core_utils.llm_cost_calc.utils import get_fedramp_uplift_multiplier + + assert get_fedramp_uplift_multiplier(model_info, data_residency) == 1.0 + + def test_service_tier_suffixes_constant_in_sync_with_enum(): from litellm.litellm_core_utils.llm_cost_calc.utils import _SERVICE_TIER_SUFFIXES from litellm.types.utils import ServiceTier diff --git a/tests/unit/litellm_core_utils/test_get_litellm_params.py b/tests/unit/litellm_core_utils/test_get_litellm_params.py index 738ab774b2f..a1184039102 100644 --- a/tests/unit/litellm_core_utils/test_get_litellm_params.py +++ b/tests/unit/litellm_core_utils/test_get_litellm_params.py @@ -340,6 +340,13 @@ class TestGetLitellmParamsDataResidency: ) assert result["data_residency"] == "us" + def test_fedramp_host_resolves_to_fedramp(self): + result = get_litellm_params( + custom_llm_provider="openai", + api_base="https://gov.api.openai.com/v1", + ) + assert result["data_residency"] == "fedramp" + def test_global_host_resolves_to_none(self): result = get_litellm_params( custom_llm_provider="openai", diff --git a/tests/unit/llms/openai/responses/test_openai_responses_data_residency.py b/tests/unit/llms/openai/responses/test_openai_responses_data_residency.py index cf9f9706cfe..8fce7a73422 100644 --- a/tests/unit/llms/openai/responses/test_openai_responses_data_residency.py +++ b/tests/unit/llms/openai/responses/test_openai_responses_data_residency.py @@ -112,6 +112,28 @@ def test_responses_us_api_base_sets_data_residency(): assert logging_obj.litellm_params.get("data_residency") == "us" +def test_responses_fedramp_api_base_sets_data_residency(): + mock_client = _make_mock_http_client(_make_responses_api_response_body()) + captured, init_spy = _capture_logging_obj() + + with ( + patch( + "litellm.llms.custom_httpx.llm_http_handler.get_httpx_client", + return_value=mock_client, + ), + patch.object(litellm.Logging, "__init__", init_spy), + ): + litellm.responses( + model="gpt-4.1", + input="hi", + api_base="https://gov.api.openai.com/v1", + api_key="test-key", + ) + + logging_obj = captured["logging_obj"] + assert logging_obj.litellm_params.get("data_residency") == "fedramp" + + def test_responses_global_api_base_leaves_data_residency_none(): mock_client = _make_mock_http_client(_make_responses_api_response_body()) captured, init_spy = _capture_logging_obj() diff --git a/tests/unit/llms/openai/test_data_residency.py b/tests/unit/llms/openai/test_data_residency.py index ecb5739133c..443d14d3102 100644 --- a/tests/unit/llms/openai/test_data_residency.py +++ b/tests/unit/llms/openai/test_data_residency.py @@ -2,7 +2,9 @@ import pytest +import litellm from litellm.llms.openai.data_residency import infer_openai_data_residency +from litellm.types.utils import DataResidency @pytest.mark.parametrize( @@ -13,6 +15,8 @@ from litellm.llms.openai.data_residency import infer_openai_data_residency ("https://us.api.openai.com/v1", "us"), ("https://us.api.openai.com", "us"), ("https://EU.api.openai.com/v1", "eu"), + ("https://gov.api.openai.com/v1", DataResidency.FEDRAMP.value), + ("https://GOV.api.openai.com/v1", DataResidency.FEDRAMP.value), ("https://api.openai.com/v1", None), ("https://api.openai.com", None), ("https://example.com/v1", None), @@ -28,7 +32,26 @@ def test_infer_openai_data_residency(api_base, expected): @pytest.mark.parametrize("custom_llm_provider", [None, "anthropic", "azure", "bedrock"]) def test_infer_openai_data_residency_non_openai_provider(custom_llm_provider): - assert ( - infer_openai_data_residency(custom_llm_provider, "https://eu.api.openai.com/v1") - is None - ) + assert infer_openai_data_residency(custom_llm_provider, "https://eu.api.openai.com/v1") is None + + +@pytest.mark.parametrize("env_var", ["OPENAI_BASE_URL", "OPENAI_API_BASE"]) +def test_infer_openai_data_residency_from_env_when_api_base_unset(monkeypatch, env_var): + monkeypatch.setattr(litellm, "api_base", None) + monkeypatch.delenv("OPENAI_BASE_URL", raising=False) + monkeypatch.delenv("OPENAI_API_BASE", raising=False) + monkeypatch.setenv(env_var, "https://gov.api.openai.com/v1") + assert infer_openai_data_residency("openai", None) == DataResidency.FEDRAMP.value + + +def test_infer_openai_data_residency_explicit_api_base_wins_over_env(monkeypatch): + monkeypatch.setattr(litellm, "api_base", None) + monkeypatch.setenv("OPENAI_BASE_URL", "https://gov.api.openai.com/v1") + assert infer_openai_data_residency("openai", "https://eu.api.openai.com/v1") == "eu" + + +def test_infer_openai_data_residency_from_litellm_api_base(monkeypatch): + monkeypatch.delenv("OPENAI_BASE_URL", raising=False) + monkeypatch.delenv("OPENAI_API_BASE", raising=False) + monkeypatch.setattr(litellm, "api_base", "https://us.api.openai.com/v1") + assert infer_openai_data_residency("openai", None) == "us" diff --git a/tests/unit/test_main.py b/tests/unit/test_main.py index 47629aab326..0220495032b 100644 --- a/tests/unit/test_main.py +++ b/tests/unit/test_main.py @@ -11547,3 +11547,28 @@ async def test_bedrock_strict_tools_limit_rejection_is_not_sent_again_when_no_to assert error is not None and "Too many strict tools" in str(error) assert len(sent) == 1 + + +def test_supplied_openai_client_base_url_decides_data_residency(monkeypatch: pytest.MonkeyPatch) -> None: + """A caller-supplied OpenAI client sends the request to its own base_url, so billing must use that + host and not an env-configured FedRAMP or regional base.""" + import openai + + monkeypatch.setattr(litellm, "api_base", None) + monkeypatch.setenv("OPENAI_BASE_URL", "https://gov.api.openai.com/v1") + captured = {} + real_init = litellm.Logging.__init__ + + def init_spy(self, *args, **kwargs): + real_init(self, *args, **kwargs) + captured["logging_obj"] = self + + with patch.object(litellm.Logging, "__init__", init_spy): + litellm.completion( + model="gpt-4.1", + messages=[{"role": "user", "content": "hi"}], + client=openai.OpenAI(api_key="test-key", base_url="https://api.openai.com/v1"), + mock_response="ok", + ) + + assert captured["logging_obj"].litellm_params.get("data_residency") is None diff --git a/tests/unit/test_utils.py b/tests/unit/test_utils.py index 32063f65765..e293b2a5be3 100644 --- a/tests/unit/test_utils.py +++ b/tests/unit/test_utils.py @@ -1079,6 +1079,7 @@ _COST_MAP_ROW_SCHEMA: Final = { "regional_endpoint_uplift_multiplier": {"type": "number"}, "regional_processing_uplift_multiplier_eu": {"type": "number"}, "regional_processing_uplift_multiplier_us": {"type": "number"}, + "fedramp_uplift_multiplier": {"type": "number"}, "input_cost_per_pixel": {"type": "number"}, "input_cost_per_query": {"type": "number"}, "input_cost_per_request": {"type": "number"}, diff --git a/ui/litellm-dashboard/src/lib/http/schema.d.ts b/ui/litellm-dashboard/src/lib/http/schema.d.ts index 068d48564ec..1963a0302c5 100644 --- a/ui/litellm-dashboard/src/lib/http/schema.d.ts +++ b/ui/litellm-dashboard/src/lib/http/schema.d.ts @@ -35973,6 +35973,8 @@ export interface components { drop_params?: boolean | string | null; /** Explicit Cache Read Input Token Cost */ explicit_cache_read_input_token_cost?: number | null; + /** Fedramp Uplift Multiplier */ + fedramp_uplift_multiplier?: number | null; /** * Fireworks Forward User Id * @description Send the LiteLLM user id of the calling key as the `user` field on Fireworks AI chat, responses and messages requests. @@ -51677,6 +51679,8 @@ export interface components { drop_params?: boolean | string | null; /** Explicit Cache Read Input Token Cost */ explicit_cache_read_input_token_cost?: number | null; + /** Fedramp Uplift Multiplier */ + fedramp_uplift_multiplier?: number | null; /** * Fireworks Forward User Id * @description Send the LiteLLM user id of the calling key as the `user` field on Fireworks AI chat, responses and messages requests.