diff --git a/src/backend/src/services/ai/chat/providers/MistralAiProvider/models.ts b/src/backend/src/services/ai/chat/providers/MistralAiProvider/models.ts index 279bdcbed1..a0929edb9b 100644 --- a/src/backend/src/services/ai/chat/providers/MistralAiProvider/models.ts +++ b/src/backend/src/services/ai/chat/providers/MistralAiProvider/models.ts @@ -16,6 +16,7 @@ export const MISTRAL_MODELS: IChatModel[] = [ 'mistral-medium', 'mistralai/mistral-medium-2508', ], + context: 131072, max_tokens: 131072, description: 'Update on Mistral Medium 3 with improved capabilities.', provider: 'mistral', @@ -42,6 +43,7 @@ export const MISTRAL_MODELS: IChatModel[] = [ 'mistral-tiny-2312', 'mistralai/open-mistral-7b', ], + context: 32768, max_tokens: 32768, description: 'Our first dense model released September 2023.', provider: 'mistral', @@ -69,6 +71,7 @@ export const MISTRAL_MODELS: IChatModel[] = [ 'mistral-tiny-latest', 'mistralai/open-mistral-nemo', ], + context: 131072, max_tokens: 131072, description: 'Our best multilingual open source model released July 2024.', provider: 'mistral', @@ -95,6 +98,7 @@ export const MISTRAL_MODELS: IChatModel[] = [ 'mistral-large-pixtral-2411', 'mistralai/pixtral-large-2411', ], + context: 131072, max_tokens: 131072, description: 'Official pixtral-large-2411 Mistral AI model', provider: 'mistral', @@ -120,6 +124,7 @@ export const MISTRAL_MODELS: IChatModel[] = [ 'codestral-latest', 'mistralai/codestral-2508', ], + context: 256000, max_tokens: 256000, description: 'Our cutting-edge language model for coding released August 2025.', provider: 'mistral', @@ -145,6 +150,7 @@ export const MISTRAL_MODELS: IChatModel[] = [ 'devstral-small-latest', 'mistralai/devstral-small-2507', ], + context: 131072, max_tokens: 131072, description: 'Our small open-source code-agentic model.', provider: 'mistral', @@ -171,6 +177,7 @@ export const MISTRAL_MODELS: IChatModel[] = [ 'devstral-medium-latest', 'mistralai/devstral-medium-2507', ], + context: 131072, max_tokens: 131072, description: 'Our medium code-agentic model.', provider: 'mistral', @@ -197,6 +204,7 @@ export const MISTRAL_MODELS: IChatModel[] = [ 'mistral-small-latest', 'mistralai/mistral-small-2506', ], + context: 131072, max_tokens: 131072, description: 'Our latest enterprise-grade small model with the latest version released June 2025.', provider: 'mistral', @@ -222,6 +230,7 @@ export const MISTRAL_MODELS: IChatModel[] = [ 'magistral-medium-latest', 'mistralai/magistral-medium-2509', ], + context: 131072, max_tokens: 131072, description: 'Our frontier-class reasoning model release candidate September 2025.', provider: 'mistral', @@ -247,6 +256,7 @@ export const MISTRAL_MODELS: IChatModel[] = [ 'magistral-small-latest', 'mistralai/magistral-small-2509', ], + context: 131072, max_tokens: 131072, description: 'Our efficient reasoning model released September 2025.', provider: 'mistral', @@ -267,6 +277,7 @@ export const MISTRAL_MODELS: IChatModel[] = [ 'voxtral-mini-latest', 'mistralai/voxtral-mini-2507', ], + context: 32768, max_tokens: 32768, description: 'A mini audio understanding model released in July 2025', provider: 'mistral', @@ -287,6 +298,7 @@ export const MISTRAL_MODELS: IChatModel[] = [ 'voxtral-small-latest', 'mistralai/voxtral-small-2507', ], + context: 32768, max_tokens: 32768, description: 'A small audio understanding model released in July 2025', provider: 'mistral', @@ -312,6 +324,7 @@ export const MISTRAL_MODELS: IChatModel[] = [ 'mistral-large-2512', 'mistralai/mistral-large-2512', ], + context: 262144, max_tokens: 262144, description: 'Official mistral-large-2512 Mistral AI model', provider: 'mistral', @@ -337,6 +350,7 @@ export const MISTRAL_MODELS: IChatModel[] = [ 'ministral-3b-latest', 'mistralai/ministral-3b-2512', ], + context: 131072, max_tokens: 131072, description: 'Ministral 3 (a.k.a. Tinystral) 3B Instruct.', provider: 'mistral', @@ -362,6 +376,7 @@ export const MISTRAL_MODELS: IChatModel[] = [ 'ministral-8b-latest', 'mistralai/ministral-8b-2512', ], + context: 262144, max_tokens: 262144, description: 'Ministral 3 (a.k.a. Tinystral) 8B Instruct.', provider: 'mistral', @@ -382,6 +397,7 @@ export const MISTRAL_MODELS: IChatModel[] = [ 'ministral-14b-latest', 'mistralai/ministral-14b-2512', ], + context: 262144, max_tokens: 262144, description: 'Ministral 3 (a.k.a. Tinystral) 14B Instruct.', provider: 'mistral', diff --git a/src/backend/src/services/ai/chat/providers/OpenAiProvider/models.ts b/src/backend/src/services/ai/chat/providers/OpenAiProvider/models.ts index f64a4b71aa..227cc65f71 100644 --- a/src/backend/src/services/ai/chat/providers/OpenAiProvider/models.ts +++ b/src/backend/src/services/ai/chat/providers/OpenAiProvider/models.ts @@ -22,7 +22,8 @@ export const OPEN_AI_MODELS: IChatModel[] = [ cached_tokens: 25, completion_tokens: 1500, }, - max_tokens: 1_050_000, // this is used for context length calculations so its misnamed from when OpenAI max_tokens and content_length were the same value + context: 1_050_000, + max_tokens: 1_050_000, }, { puterId: 'openai:openai/gpt-5.3-codex', @@ -41,6 +42,7 @@ export const OPEN_AI_MODELS: IChatModel[] = [ cached_tokens: 17.5, completion_tokens: 1400, }, + context: 128_000, max_tokens: 128000, responses_api_only: true, }, @@ -62,6 +64,7 @@ export const OPEN_AI_MODELS: IChatModel[] = [ cached_tokens: 18, completion_tokens: 1400, }, + context: 128_000, max_tokens: 128000, responses_api_only: true, }, @@ -83,6 +86,7 @@ export const OPEN_AI_MODELS: IChatModel[] = [ cached_tokens: 17.5, completion_tokens: 1400, }, + context: 128_000, max_tokens: 16384, }, { @@ -102,6 +106,7 @@ export const OPEN_AI_MODELS: IChatModel[] = [ prompt_tokens: 2100, completion_tokens: 16800, }, + context: 128_000, max_tokens: 16384, responses_api_only: true, }, @@ -123,6 +128,7 @@ export const OPEN_AI_MODELS: IChatModel[] = [ cached_tokens: 17.5, completion_tokens: 1400, }, + context: 128_000, max_tokens: 128000, }, { @@ -143,6 +149,7 @@ export const OPEN_AI_MODELS: IChatModel[] = [ cached_tokens: 13, completion_tokens: 1000, }, + context: 128_000, max_tokens: 128000, }, { @@ -163,6 +170,7 @@ export const OPEN_AI_MODELS: IChatModel[] = [ cached_tokens: 13, completion_tokens: 1000, }, + context: 128_000, max_tokens: 128000, responses_api_only: true, }, @@ -184,6 +192,7 @@ export const OPEN_AI_MODELS: IChatModel[] = [ cached_tokens: 3, completion_tokens: 200, }, + context: 128_000, max_tokens: 128000, responses_api_only: true, }, @@ -205,6 +214,7 @@ export const OPEN_AI_MODELS: IChatModel[] = [ cached_tokens: 13, completion_tokens: 1000, }, + context: 128_000, max_tokens: 16384, }, { @@ -225,6 +235,7 @@ export const OPEN_AI_MODELS: IChatModel[] = [ cached_tokens: 13, completion_tokens: 1000, }, + context: 128_000, max_tokens: 128000, }, { @@ -245,6 +256,7 @@ export const OPEN_AI_MODELS: IChatModel[] = [ cached_tokens: 3, completion_tokens: 200, }, + context: 128_000, max_tokens: 128000, }, { @@ -265,6 +277,7 @@ export const OPEN_AI_MODELS: IChatModel[] = [ cached_tokens: 1, completion_tokens: 40, }, + context: 128_000, max_tokens: 128000, }, { @@ -285,6 +298,7 @@ export const OPEN_AI_MODELS: IChatModel[] = [ cached_tokens: 13, completion_tokens: 1000, }, + context: 128_000, max_tokens: 16384, }, { @@ -305,6 +319,7 @@ export const OPEN_AI_MODELS: IChatModel[] = [ cached_tokens: 125, completion_tokens: 1000, }, + context: 128_000, max_tokens: 16384, }, { @@ -316,6 +331,7 @@ export const OPEN_AI_MODELS: IChatModel[] = [ knowledge: '2023-09', release_date: '2024-07-18', aliases: ['openai/gpt-4o-mini'], + context: 128_000, max_tokens: 16384, costs_currency: 'usd-cents', input_cost_key: 'prompt_tokens', @@ -345,6 +361,7 @@ export const OPEN_AI_MODELS: IChatModel[] = [ cached_tokens: 750, completion_tokens: 6000, }, + context: 200_000, max_tokens: 100000, }, { @@ -364,6 +381,7 @@ export const OPEN_AI_MODELS: IChatModel[] = [ prompt_tokens: 110, completion_tokens: 440, }, + context: 128_000, max_tokens: 65536, }, { @@ -383,6 +401,7 @@ export const OPEN_AI_MODELS: IChatModel[] = [ prompt_tokens: 15000, completion_tokens: 60000, }, + context: 200_000, max_tokens: 100000, }, { @@ -403,6 +422,7 @@ export const OPEN_AI_MODELS: IChatModel[] = [ cached_tokens: 50, completion_tokens: 800, }, + context: 200_000, max_tokens: 100000, }, { @@ -423,6 +443,7 @@ export const OPEN_AI_MODELS: IChatModel[] = [ cached_tokens: 50, completion_tokens: 8000, }, + context: 200_000, max_tokens: 100000, responses_api_only: true, }, @@ -444,6 +465,7 @@ export const OPEN_AI_MODELS: IChatModel[] = [ cached_tokens: 55, completion_tokens: 440, }, + context: 200_000, max_tokens: 100000, }, { @@ -463,6 +485,7 @@ export const OPEN_AI_MODELS: IChatModel[] = [ prompt_tokens: 110, completion_tokens: 440, }, + context: 200_000, max_tokens: 100000, }, { @@ -483,6 +506,7 @@ export const OPEN_AI_MODELS: IChatModel[] = [ cached_tokens: 50, completion_tokens: 800, }, + context: 1_047_576, max_tokens: 32768, }, { @@ -503,6 +527,7 @@ export const OPEN_AI_MODELS: IChatModel[] = [ cached_tokens: 10, completion_tokens: 160, }, + context: 1_047_576, max_tokens: 32768, }, { @@ -523,6 +548,7 @@ export const OPEN_AI_MODELS: IChatModel[] = [ cached_tokens: 2, completion_tokens: 40, }, + context: 1_047_576, max_tokens: 32768, }, { @@ -537,6 +563,7 @@ export const OPEN_AI_MODELS: IChatModel[] = [ prompt_tokens: 7500, completion_tokens: 15000, }, + context: 128_000, max_tokens: 32768, }, ]; diff --git a/src/backend/src/services/ai/chat/providers/OpenRouterProvider/OpenRouterProvider.ts b/src/backend/src/services/ai/chat/providers/OpenRouterProvider/OpenRouterProvider.ts index b12ed90a76..57685c822b 100644 --- a/src/backend/src/services/ai/chat/providers/OpenRouterProvider/OpenRouterProvider.ts +++ b/src/backend/src/services/ai/chat/providers/OpenRouterProvider/OpenRouterProvider.ts @@ -186,6 +186,7 @@ export class OpenRouterProvider implements IChatProvider { id: `openrouter:${model.id}`, name: `${model.name} (OpenRouter)`, aliases: [model.id, model.name, `openrouter/${model.id}`, model.id.split('/').slice(1).join('/')], + context: model.context_length, max_tokens: model.top_provider.max_completion_tokens, costs_currency: 'usd-cents', input_cost_key: 'prompt',