desirecore/config-center
2
0
Fork 0
mirror of https://git.openapi.site/https://github.com/desirecore/config-center.git synced 2026-09-05 19:13:33 +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

208
compute/model-specs/zhipu.json
View File

@@ -5,14 +5,48 @@
"id": "glm-5.2",
"displayName": "GLM-5.2",
"family": "glm-5.2",
"match": { "exact": ["glm-5.2"], "patterns": ["glm-5.2*"] },
"match": {
"exact": [
"glm-5.2"
],
"patterns": [
"glm-5.2*"
]
},
"spec": {
"contextWindow": 1048576,
"maxOutputTokens": 32768,
"capabilities": ["reasoning", "deep_thinking", "code", "multilingual", "tool_use", "long_context"],
"serviceType": ["chat"],
"capabilities": [
"reasoning",
"deep_thinking",
"code",
"multilingual",
"tool_use",
"long_context"
],
"serviceType": [
"chat"
],
"supportsReasoning": true,
"description": "智谱 GLM-5.2 大推理模型，100 万上下文"
},
"routing": {
"tier": "flagship",
"routingPriority": 45,
"eligibleForAgent": true,
"defaultReference": false,
"reasoning": {
"supportedModes": [
"auto",
"off",
"minimal",
"low",
"medium",
"high",
"xhigh"
],
"defaultMode": "high"
}
}
},
{
@@ -22,10 +56,40 @@
"spec": {
"contextWindow": 200000,
"maxOutputTokens": 128000,
"capabilities": ["chat", "reasoning", "code", "multilingual", "deep_thinking", "long_context", "math", "tool_use", "agent"],
"serviceType": ["chat"],
"capabilities": [
"chat",
"reasoning",
"code",
"multilingual",
"deep_thinking",
"long_context",
"math",
"tool_use",
"agent"
],
"serviceType": [
"chat"
],
"defaultTemperature": 1,
"supportsReasoning": true
},
"routing": {
"tier": "balanced",
"routingPriority": 65,
"eligibleForAgent": true,
"defaultReference": false,
"reasoning": {
"supportedModes": [
"auto",
"off",
"minimal",
"low",
"medium",
"high",
"xhigh"
],
"defaultMode": "medium"
}
}
},
{
@@ -35,8 +99,20 @@
"spec": {
"contextWindow": 200000,
"maxOutputTokens": 128000,
"capabilities": ["chat", "reasoning", "code", "multilingual", "deep_thinking", "long_context", "math", "tool_use", "agent"],
"serviceType": ["chat"],
"capabilities": [
"chat",
"reasoning",
"code",
"multilingual",
"deep_thinking",
"long_context",
"math",
"tool_use",
"agent"
],
"serviceType": [
"chat"
],
"defaultTemperature": 1,
"supportsReasoning": true
}
@@ -48,8 +124,18 @@
"spec": {
"contextWindow": 200000,
"maxOutputTokens": 128000,
"capabilities": ["chat", "reasoning", "code", "deep_thinking", "long_context", "tool_use", "agent"],
"serviceType": ["chat"],
"capabilities": [
"chat",
"reasoning",
"code",
"deep_thinking",
"long_context",
"tool_use",
"agent"
],
"serviceType": [
"chat"
],
"defaultTemperature": 1,
"supportsReasoning": true
}
@@ -61,8 +147,18 @@
"spec": {
"contextWindow": 200000,
"maxOutputTokens": 128000,
"capabilities": ["chat", "reasoning", "code", "multilingual", "deep_thinking", "long_context", "tool_use"],
"serviceType": ["chat"],
"capabilities": [
"chat",
"reasoning",
"code",
"multilingual",
"deep_thinking",
"long_context",
"tool_use"
],
"serviceType": [
"chat"
],
"defaultTemperature": 1,
"supportsReasoning": true
}
@@ -74,8 +170,16 @@
"spec": {
"contextWindow": 200000,
"maxOutputTokens": 128000,
"capabilities": ["chat", "reasoning", "code", "multilingual", "deep_thinking"],
"serviceType": ["chat"],
"capabilities": [
"chat",
"reasoning",
"code",
"multilingual",
"deep_thinking"
],
"serviceType": [
"chat"
],
"defaultTemperature": 1,
"supportsReasoning": true
}
@@ -84,12 +188,24 @@
"id": "glm-4.7-thinking",
"displayName": "GLM-4.7 Thinking",
"family": "glm-4.7",
"match": { "exact": ["glm-4.7-thinking"] },
"match": {
"exact": [
"glm-4.7-thinking"
]
},
"spec": {
"contextWindow": 200000,
"maxOutputTokens": 128000,
"capabilities": ["reasoning", "math", "code", "deep_thinking", "long_context"],
"serviceType": ["reasoning"],
"capabilities": [
"reasoning",
"math",
"code",
"deep_thinking",
"long_context"
],
"serviceType": [
"reasoning"
],
"defaultTemperature": null,
"supportsReasoning": true,
"description": "智谱GLM-4.7深度思考模式，交错式/保留式/轮级思考"
@@ -99,12 +215,30 @@
"id": "glm-5v-turbo",
"displayName": "GLM-5V-Turbo",
"family": "glm-5v",
"match": { "exact": ["glm-5v-turbo"] },
"match": {
"exact": [
"glm-5v-turbo"
]
},
"spec": {
"contextWindow": 200000,
"maxOutputTokens": 128000,
"capabilities": ["chat", "vision", "video_understanding", "image_understanding", "file_understanding", "reasoning", "code", "deep_thinking", "long_context", "tool_use", "agent"],
"serviceType": ["vision"],
"capabilities": [
"chat",
"vision",
"video_understanding",
"image_understanding",
"file_understanding",
"reasoning",
"code",
"deep_thinking",
"long_context",
"tool_use",
"agent"
],
"serviceType": [
"vision"
],
"defaultTemperature": 1,
"supportsReasoning": true,
"description": "智谱首个多模态 Coding 基座模型，支持视频、图像、文本和文件输入"
@@ -114,12 +248,25 @@
"id": "glm-4.6v",
"displayName": "GLM-4.6V",
"family": "glm-4.6",
"match": { "exact": ["glm-4.6v"] },
"match": {
"exact": [
"glm-4.6v"
]
},
"spec": {
"contextWindow": 128000,
"maxOutputTokens": 32768,
"capabilities": ["chat", "vision", "video_understanding", "image_understanding", "long_context", "tool_use"],
"serviceType": ["vision"],
"capabilities": [
"chat",
"vision",
"video_understanding",
"image_understanding",
"long_context",
"tool_use"
],
"serviceType": [
"vision"
],
"defaultTemperature": 1,
"description": "智谱GLM-4.6V多模态版，106B/12B MoE，支持图像视频理解"
}
@@ -128,11 +275,22 @@
"id": "embedding-3",
"displayName": "智谱 embedding-3",
"family": "zhipu-embedding",
"match": { "exact": ["embedding-3"] },
"match": {
"exact": [
"embedding-3"
]
},
"spec": {
"contextWindow": 8192,
"capabilities": ["text_embedding", "semantic_search", "rag", "custom_dimensions"],
"serviceType": ["embedding"],
"capabilities": [
"text_embedding",
"semantic_search",
"rag",
"custom_dimensions"
],
"serviceType": [
"embedding"
],
"description": "智谱嵌入模型v3，支持自定义维度；单条输入最多 3072 tokens"
}
}