desirecore/config-center
2
0
Fork 0
mirror of https://git.openapi.site/https://github.com/desirecore/config-center.git synced 2026-09-05 17:53:41 +08:00
Code Issues Packages Projects Releases Wiki Activity

refactor: consolidate smart routing into model specs (#75)

Browse Source
This commit is contained in:
yi-ge
2026-08-10 19:13:08 +08:00
committed by GitHub
parent ed0aeccf29
commit ad8c515938
32 changed files with 3907 additions and 2021 deletions
Download Patch File Download Diff File Expand all files Collapse all files

682
compute/model-specs/openai.json
View File

@@ -5,11 +5,24 @@
"id": "gpt-image-2",
"displayName": "GPT Image 2",
"family": "gpt-image",
"match": { "exact": ["gpt-image-2"], "patterns": ["gpt-image-2*"] },
"match": {
"exact": [
"gpt-image-2"
],
"patterns": [
"gpt-image-2*"
]
},
"spec": {
"contextWindow": 400000,
"capabilities": ["image_generation", "image_editing", "high_quality"],
"serviceType": ["image_gen"],
"capabilities": [
"image_generation",
"image_editing",
"high_quality"
],
"serviceType": [
"image_gen"
],
"description": "OpenAI 最新文生图模型，支持高质量图片生成与编辑，400K 上下文"
}
},
@@ -17,28 +30,94 @@
"id": "gpt-5.5",
"displayName": "GPT-5.5",
"family": "gpt-5",
"match": { "exact": ["gpt-5.5"], "patterns": ["gpt-5.5*"] },
"match": {
"exact": [
"gpt-5.5"
],
"patterns": [
"gpt-5.5*"
]
},
"spec": {
"contextWindow": 1050000,
"maxOutputTokens": 128000,
"capabilities": ["chat", "reasoning", "code", "vision", "long_context", "tool_use"],
"serviceType": ["chat"],
"capabilities": [
"chat",
"reasoning",
"code",
"vision",
"long_context",
"tool_use"
],
"serviceType": [
"chat"
],
"defaultTemperature": 1,
"supportsReasoning": true
},
"routing": {
"tier": "flagship",
"routingPriority": 60,
"eligibleForAgent": true,
"defaultReference": false,
"reasoning": {
"supportedModes": [
"auto",
"off",
"low",
"medium",
"high",
"xhigh"
],
"defaultMode": "medium"
}
}
},
{
"id": "gpt-5.4",
"displayName": "GPT-5.4",
"family": "gpt-5",
"match": { "exact": ["gpt-5.4"], "patterns": ["gpt-5.4*"] },
"match": {
"exact": [
"gpt-5.4"
],
"patterns": [
"gpt-5.4*"
]
},
"spec": {
"contextWindow": 1050000,
"maxOutputTokens": 128000,
"capabilities": ["chat", "reasoning", "code", "vision", "long_context", "tool_use"],
"serviceType": ["chat"],
"capabilities": [
"chat",
"reasoning",
"code",
"vision",
"long_context",
"tool_use"
],
"serviceType": [
"chat"
],
"defaultTemperature": 1,
"supportsReasoning": true
},
"routing": {
"tier": "balanced",
"routingPriority": 60,
"eligibleForAgent": true,
"defaultReference": false,
"reasoning": {
"supportedModes": [
"auto",
"off",
"low",
"medium",
"high",
"xhigh"
],
"defaultMode": "medium"
}
}
},
{
@@ -48,8 +127,17 @@
"spec": {
"contextWindow": 400000,
"maxOutputTokens": 128000,
"capabilities": ["chat", "reasoning", "code", "vision", "long_context", "tool_use"],
"serviceType": ["chat"],
"capabilities": [
"chat",
"reasoning",
"code",
"vision",
"long_context",
"tool_use"
],
"serviceType": [
"chat"
],
"defaultTemperature": 1,
"supportsReasoning": true
}
@@ -61,8 +149,17 @@
"spec": {
"contextWindow": 400000,
"maxOutputTokens": 128000,
"capabilities": ["chat", "reasoning", "code", "vision", "long_context", "tool_use"],
"serviceType": ["chat"],
"capabilities": [
"chat",
"reasoning",
"code",
"vision",
"long_context",
"tool_use"
],
"serviceType": [
"chat"
],
"defaultTemperature": 1,
"supportsReasoning": true
}
@@ -74,8 +171,17 @@
"spec": {
"contextWindow": 400000,
"maxOutputTokens": 128000,
"capabilities": ["chat", "reasoning", "code", "vision", "long_context", "tool_use"],
"serviceType": ["chat"],
"capabilities": [
"chat",
"reasoning",
"code",
"vision",
"long_context",
"tool_use"
],
"serviceType": [
"chat"
],
"defaultTemperature": 1,
"supportsReasoning": true
}
@@ -84,12 +190,24 @@
"id": "gpt-5-mini",
"displayName": "GPT-5 Mini",
"family": "gpt-5",
"match": { "patterns": ["gpt-5-mini*"] },
"match": {
"patterns": [
"gpt-5-mini*"
]
},
"spec": {
"contextWindow": 400000,
"maxOutputTokens": 128000,
"capabilities": ["chat", "fast", "tool_use", "code", "vision"],
"serviceType": ["chat"],
"capabilities": [
"chat",
"fast",
"tool_use",
"code",
"vision"
],
"serviceType": [
"chat"
],
"defaultTemperature": 1,
"supportsReasoning": true
}
@@ -98,12 +216,22 @@
"id": "gpt-5-nano",
"displayName": "GPT-5 Nano",
"family": "gpt-5",
"match": { "patterns": ["gpt-5-nano*"] },
"match": {
"patterns": [
"gpt-5-nano*"
]
},
"spec": {
"contextWindow": 400000,
"maxOutputTokens": 128000,
"capabilities": ["chat", "code", "fast"],
"serviceType": ["fast"],
"capabilities": [
"chat",
"code",
"fast"
],
"serviceType": [
"fast"
],
"defaultTemperature": 1
}
},
@@ -114,8 +242,17 @@
"spec": {
"contextWindow": 1047576,
"maxOutputTokens": 32768,
"capabilities": ["chat", "reasoning", "code", "vision", "ultra_long_context", "tool_use"],
"serviceType": ["chat"],
"capabilities": [
"chat",
"reasoning",
"code",
"vision",
"ultra_long_context",
"tool_use"
],
"serviceType": [
"chat"
],
"defaultTemperature": 1
}
},
@@ -126,8 +263,15 @@
"spec": {
"contextWindow": 128000,
"maxOutputTokens": 16384,
"capabilities": ["chat", "vision", "image_understanding", "tool_use"],
"serviceType": ["vision"],
"capabilities": [
"chat",
"vision",
"image_understanding",
"tool_use"
],
"serviceType": [
"vision"
],
"defaultTemperature": 1
}
},
@@ -135,12 +279,23 @@
"id": "gpt-4o-mini",
"displayName": "GPT-4o Mini",
"family": "gpt-4o",
"match": { "patterns": ["gpt-4o-mini*"] },
"match": {
"patterns": [
"gpt-4o-mini*"
]
},
"spec": {
"contextWindow": 128000,
"maxOutputTokens": 16384,
"capabilities": ["chat", "code", "vision", "tool_use"],
"serviceType": ["vision"],
"capabilities": [
"chat",
"code",
"vision",
"tool_use"
],
"serviceType": [
"vision"
],
"defaultTemperature": 1
}
},
@@ -151,8 +306,17 @@
"spec": {
"contextWindow": 200000,
"maxOutputTokens": 100000,
"capabilities": ["reasoning", "deep_thinking", "code", "math", "science", "tool_use"],
"serviceType": ["reasoning"],
"capabilities": [
"reasoning",
"deep_thinking",
"code",
"math",
"science",
"tool_use"
],
"serviceType": [
"reasoning"
],
"defaultTemperature": null,
"supportsReasoning": true
}
@@ -161,12 +325,23 @@
"id": "o3-mini",
"displayName": "o3 Mini",
"family": "o3",
"match": { "patterns": ["o3-mini*"] },
"match": {
"patterns": [
"o3-mini*"
]
},
"spec": {
"contextWindow": 200000,
"maxOutputTokens": 100000,
"capabilities": ["reasoning", "math", "code", "fast"],
"serviceType": ["reasoning"],
"capabilities": [
"reasoning",
"math",
"code",
"fast"
],
"serviceType": [
"reasoning"
],
"defaultTemperature": null,
"supportsReasoning": true
}
@@ -175,12 +350,24 @@
"id": "o4-mini",
"displayName": "o4 Mini",
"family": "o4",
"match": { "patterns": ["o4-mini*"] },
"match": {
"patterns": [
"o4-mini*"
]
},
"spec": {
"contextWindow": 200000,
"maxOutputTokens": 100000,
"capabilities": ["reasoning", "math", "code", "tool_use", "vision"],
"serviceType": ["reasoning"],
"capabilities": [
"reasoning",
"math",
"code",
"tool_use",
"vision"
],
"serviceType": [
"reasoning"
],
"defaultTemperature": null,
"supportsReasoning": true
}
@@ -189,12 +376,25 @@
"id": "gpt-5.2-pro",
"displayName": "GPT-5.2 Pro",
"family": "gpt-5",
"match": { "exact": ["gpt-5.2-pro"] },
"match": {
"exact": [
"gpt-5.2-pro"
]
},
"spec": {
"contextWindow": 400000,
"maxOutputTokens": 128000,
"capabilities": ["chat", "reasoning", "code", "vision", "long_context", "tool_use"],
"serviceType": ["chat"],
"capabilities": [
"chat",
"reasoning",
"code",
"vision",
"long_context",
"tool_use"
],
"serviceType": [
"chat"
],
"defaultTemperature": 1,
"supportsReasoning": true,
"description": "OpenAI GPT-5.2 Pro 专业版，最高性能"
@@ -204,12 +404,25 @@
"id": "gpt-5-pro",
"displayName": "GPT-5 Pro",
"family": "gpt-5",
"match": { "exact": ["gpt-5-pro"] },
"match": {
"exact": [
"gpt-5-pro"
]
},
"spec": {
"contextWindow": 400000,
"maxOutputTokens": 272000,
"capabilities": ["chat", "reasoning", "code", "vision", "long_context", "tool_use"],
"serviceType": ["chat"],
"capabilities": [
"chat",
"reasoning",
"code",
"vision",
"long_context",
"tool_use"
],
"serviceType": [
"chat"
],
"defaultTemperature": 1,
"supportsReasoning": true,
"description": "OpenAI GPT-5 Pro 专业版高性能推理模型"
@@ -219,12 +432,27 @@
"id": "gpt-4.1-mini",
"displayName": "GPT-4.1 mini",
"family": "gpt-4.1",
"match": { "exact": ["gpt-4.1-mini"], "patterns": ["gpt-4.1-mini*"] },
"match": {
"exact": [
"gpt-4.1-mini"
],
"patterns": [
"gpt-4.1-mini*"
]
},
"spec": {
"contextWindow": 1047576,
"maxOutputTokens": 32768,
"capabilities": ["chat", "code", "vision", "ultra_long_context", "tool_use"],
"serviceType": ["chat"],
"capabilities": [
"chat",
"code",
"vision",
"ultra_long_context",
"tool_use"
],
"serviceType": [
"chat"
],
"defaultTemperature": 1,
"description": "OpenAI GPT-4.1 mini，1M 上下文，高性价比"
}
@@ -233,12 +461,26 @@
"id": "gpt-4.1-nano",
"displayName": "GPT-4.1 nano",
"family": "gpt-4.1",
"match": { "exact": ["gpt-4.1-nano"], "patterns": ["gpt-4.1-nano*"] },
"match": {
"exact": [
"gpt-4.1-nano"
],
"patterns": [
"gpt-4.1-nano*"
]
},
"spec": {
"contextWindow": 1047576,
"maxOutputTokens": 32768,
"capabilities": ["chat", "code", "ultra_long_context", "fast"],
"serviceType": ["fast"],
"capabilities": [
"chat",
"code",
"ultra_long_context",
"fast"
],
"serviceType": [
"fast"
],
"defaultTemperature": 1,
"description": "OpenAI GPT-4.1 nano，1M 上下文，极致性价比"
}
@@ -247,11 +489,20 @@
"id": "text-embedding-3-small",
"displayName": "text-embedding-3-small",
"family": "text-embedding-3",
"match": { "exact": ["text-embedding-3-small"] },
"match": {
"exact": [
"text-embedding-3-small"
]
},
"spec": {
"contextWindow": 8192,
"capabilities": ["text_embedding", "semantic_search"],
"serviceType": ["embedding"],
"capabilities": [
"text_embedding",
"semantic_search"
],
"serviceType": [
"embedding"
],
"description": "OpenAI 轻量文本向量化模型，1536 维"
}
},
@@ -259,11 +510,20 @@
"id": "text-embedding-3-large",
"displayName": "text-embedding-3-large",
"family": "text-embedding-3",
"match": { "exact": ["text-embedding-3-large"] },
"match": {
"exact": [
"text-embedding-3-large"
]
},
"spec": {
"contextWindow": 8192,
"capabilities": ["text_embedding", "semantic_search"],
"serviceType": ["embedding"],
"capabilities": [
"text_embedding",
"semantic_search"
],
"serviceType": [
"embedding"
],
"description": "OpenAI 高精度文本向量化模型，3072 维"
}
},
@@ -271,10 +531,18 @@
"id": "tts-1",
"displayName": "TTS-1",
"family": "tts-1",
"match": { "exact": ["tts-1"] },
"match": {
"exact": [
"tts-1"
]
},
"spec": {
"capabilities": ["tts"],
"serviceType": ["tts"],
"capabilities": [
"tts"
],
"serviceType": [
"tts"
],
"description": "OpenAI 标准语音合成"
}
},
@@ -282,10 +550,18 @@
"id": "tts-1-hd",
"displayName": "TTS-1-HD",
"family": "tts-1",
"match": { "exact": ["tts-1-hd"] },
"match": {
"exact": [
"tts-1-hd"
]
},
"spec": {
"capabilities": ["tts"],
"serviceType": ["tts"],
"capabilities": [
"tts"
],
"serviceType": [
"tts"
],
"description": "OpenAI 高清语音合成"
}
},
@@ -293,10 +569,19 @@
"id": "whisper-1",
"displayName": "Whisper",
"family": "whisper",
"match": { "exact": ["whisper-1"] },
"match": {
"exact": [
"whisper-1"
]
},
"spec": {
"capabilities": ["asr", "multilingual"],
"serviceType": ["asr"],
"capabilities": [
"asr",
"multilingual"
],
"serviceType": [
"asr"
],
"description": "OpenAI 通用语音识别"
}
},
@@ -304,12 +589,24 @@
"id": "o3-pro",
"displayName": "o3-pro",
"family": "o3",
"match": { "exact": ["o3-pro"] },
"match": {
"exact": [
"o3-pro"
]
},
"spec": {
"contextWindow": 200000,
"maxOutputTokens": 100000,
"capabilities": ["reasoning", "deep_thinking", "code", "math", "science"],
"serviceType": ["reasoning"],
"capabilities": [
"reasoning",
"deep_thinking",
"code",
"math",
"science"
],
"serviceType": [
"reasoning"
],
"defaultTemperature": null,
"supportsReasoning": true,
"description": "OpenAI o3-pro 高级推理模型，200K 上下文"
@@ -319,10 +616,19 @@
"id": "dall-e-3",
"displayName": "DALL-E 3",
"family": "dall-e",
"match": { "exact": ["dall-e-3"] },
"match": {
"exact": [
"dall-e-3"
]
},
"spec": {
"capabilities": ["image_generation", "text_rendering"],
"serviceType": ["image_gen"],
"capabilities": [
"image_generation",
"text_rendering"
],
"serviceType": [
"image_gen"
],
"description": "OpenAI 高质量文生图模型"
}
},
@@ -330,12 +636,24 @@
"id": "gpt-4o-realtime",
"displayName": "GPT-4o Realtime",
"family": "gpt-4o-realtime",
"match": { "exact": ["gpt-4o-realtime"] },
"match": {
"exact": [
"gpt-4o-realtime"
]
},
"spec": {
"contextWindow": 32000,
"maxOutputTokens": 4096,
"capabilities": ["chat", "vision", "tts", "asr", "realtime"],
"serviceType": ["omni"],
"capabilities": [
"chat",
"vision",
"tts",
"asr",
"realtime"
],
"serviceType": [
"omni"
],
"defaultTemperature": 0.7,
"description": "OpenAI 全能实时模型，支持语音 + 文本 + 视觉"
}
@@ -344,12 +662,25 @@
"id": "gpt-4o-realtime-preview",
"displayName": "GPT-4o Realtime Preview",
"family": "gpt-4o-realtime",
"match": { "exact": ["gpt-4o-realtime-preview"], "patterns": ["gpt-4o-realtime-preview*"] },
"match": {
"exact": [
"gpt-4o-realtime-preview"
],
"patterns": [
"gpt-4o-realtime-preview*"
]
},
"spec": {
"contextWindow": 32000,
"maxOutputTokens": 4096,
"capabilities": ["realtime", "speech_to_speech", "low_latency"],
"serviceType": ["realtime_voice"],
"capabilities": [
"realtime",
"speech_to_speech",
"low_latency"
],
"serviceType": [
"realtime_voice"
],
"defaultTemperature": 0.7,
"description": "OpenAI 实时语音交互模型（预览版）"
}
@@ -358,14 +689,209 @@
"id": "gpt-oss-120b",
"displayName": "GPT OSS 120B",
"family": "gpt-oss",
"match": { "exact": ["gpt-oss-120b", "openai/gpt-oss-120b:free", "gpt-oss-120b:free"] },
"match": {
"exact": [
"gpt-oss-120b",
"openai/gpt-oss-120b:free",
"gpt-oss-120b:free"
]
},
"spec": {
"contextWindow": 128000,
"maxOutputTokens": 16384,
"capabilities": ["chat", "code", "reasoning"],
"serviceType": ["chat"],
"capabilities": [
"chat",
"code",
"reasoning"
],
"serviceType": [
"chat"
],
"description": "OpenAI 开源 120B 模型（通过 OpenRouter 提供，:free 层）"
}
},
{
"id": "gpt-5.6-sol",
"displayName": "GPT-5.6 Sol",
"spec": {
"contextWindow": 1050000,
"maxOutputTokens": 128000,
"capabilities": [
"chat",
"reasoning",
"code",
"vision",
"long_context",
"tool_use",
"agent"
],
"serviceType": [
"chat"
],
"supportsReasoning": true
},
"routing": {
"tier": "flagship",
"routingPriority": 10,
"eligibleForAgent": true,
"defaultReference": false,
"reasoning": {
"supportedModes": [
"auto",
"off",
"low",
"medium",
"high",
"xhigh",
"max"
],
"defaultMode": "low"
}
}
},
{
"id": "gpt-5.6-terra",
"displayName": "GPT-5.6 Terra",
"spec": {
"contextWindow": 1050000,
"maxOutputTokens": 128000,
"capabilities": [
"chat",
"reasoning",
"code",
"vision",
"long_context",
"tool_use",
"agent"
],
"serviceType": [
"chat"
],
"supportsReasoning": true
},
"routing": {
"tier": "balanced",
"routingPriority": 10,
"eligibleForAgent": true,
"defaultReference": false,
"reasoning": {
"supportedModes": [
"auto",
"off",
"low",
"medium",
"high",
"xhigh",
"max"
],
"defaultMode": "medium"
}
}
},
{
"id": "gpt-5.6-luna",
"displayName": "GPT-5.6 Luna",
"spec": {
"contextWindow": 1050000,
"maxOutputTokens": 128000,
"capabilities": [
"chat",
"reasoning",
"code",
"vision",
"long_context",
"tool_use",
"fast"
],
"serviceType": [
"chat"
],
"supportsReasoning": true
},
"routing": {
"tier": "lightweight",
"routingPriority": 10,
"eligibleForAgent": true,
"defaultReference": false,
"reasoning": {
"supportedModes": [
"auto",
"off",
"low",
"medium",
"high",
"xhigh",
"max"
],
"defaultMode": "medium"
}
}
},
{
"id": "gpt-5.4-mini",
"displayName": "GPT-5.4 Mini",
"spec": {
"contextWindow": 400000,
"maxOutputTokens": 128000,
"capabilities": [
"chat",
"reasoning",
"code",
"vision",
"long_context",
"tool_use",
"fast"
],
"serviceType": [
"chat"
],
"supportsReasoning": true
},
"routing": {
"tier": "lightweight",
"routingPriority": 25,
"eligibleForAgent": true,
"defaultReference": false,
"reasoning": {
"supportedModes": [
"auto",
"off",
"low",
"medium",
"high",
"xhigh"
],
"defaultMode": "low"
}
}
},
{
"id": "gpt-5.3-codex-spark",
"displayName": "GPT-5.3 Codex Spark（Pro 专属）",
"spec": {
"contextWindow": 128000,
"capabilities": [
"chat",
"code",
"tool_use",
"fast"
],
"serviceType": [
"chat"
]
},
"routing": {
"tier": "lightweight",
"routingPriority": 30,
"eligibleForAgent": true,
"defaultReference": false,
"reasoning": {
"supportedModes": [
"auto"
],
"defaultMode": "auto"
}
}
}
]
}