desirecore/config-center
2
0
Fork 0
mirror of https://git.openapi.site/https://github.com/desirecore/config-center.git synced 2026-06-06 05:50:50 +08:00
Code Issues Packages Projects Releases Wiki Activity

Revert "Merge pull request #1 from desirecore/model-field-official-audit"

Browse Source 
This reverts commit 88a8cf2db6, reversing
changes made to ca99ea272f.
This commit is contained in:
yi-ge
2026-04-25 19:34:29 +08:00
parent 88a8cf2db6
commit 371336cdd6
114 changed files with 527 additions and 6091 deletions
Download Patch File Download Diff File Expand all files Collapse all files

54
compute/coding-plans/dashscope-coding.json
View File

@@ -10,10 +10,7 @@
"enabled": false,
"status": "unconfigured",
"accessMode": "coding-plan",
"services": [
"chat",
"reasoning"
],
"services": ["chat", "reasoning"],
"codingPlan": {
"quotas": {},
"usageTracking": {
@@ -25,47 +22,22 @@
{
"modelName": "qwen3-coder-plus",
"displayName": "Qwen3 Coder Plus (Coding)",
"serviceType": [
"chat"
],
"description": "阿里云百炼 Coding Plan，Qwen3 编程增强版，100万上下文",
"contextWindow": 1000000,
"maxOutputTokens": 65536,
"capabilities": [
"chat",
"reasoning",
"code",
"tool_use"
],
"source": "preset",
"inputPrice": 1,
"outputPrice": 5,
"defaultTemperature": 0.7,
"defaultTopP": 0.8
"serviceType": ["chat"],
"description": "阿里云百炼 Coding Plan，Qwen3 编程增强版",
"contextWindow": 131072,
"maxOutputTokens": 16384,
"capabilities": ["chat", "reasoning", "code", "tool_use"],
"source": "preset"
},
{
"modelName": "qwen3-max",
"displayName": "Qwen3 Max (Coding)",
"serviceType": [
"chat"
],
"description": "阿里云百炼 Coding Plan，Qwen3 Max 旗舰模型，262K 上下文",
"contextWindow": 262144,
"maxOutputTokens": 65536,
"capabilities": [
"chat",
"reasoning",
"code",
"multilingual",
"tool_use",
"long_context",
"vision"
],
"source": "preset",
"inputPrice": 1.2,
"outputPrice": 6,
"defaultTemperature": 0.7,
"defaultTopP": 0.8
"serviceType": ["chat"],
"description": "阿里云百炼 Coding Plan，Qwen3 Max 旗舰模型",
"contextWindow": 131072,
"maxOutputTokens": 16384,
"capabilities": ["chat", "reasoning", "code", "multilingual", "tool_use", "long_context", "vision"],
"source": "preset"
}
]
}

2
compute/coding-plans/infini-coding.json
View File

@@ -27,8 +27,6 @@
"contextWindow": 128000,
"maxOutputTokens": 8192,
"capabilities": ["chat", "reasoning", "code", "multilingual"],
"defaultTemperature": null,
"defaultTopP": null,
"source": "preset"
}
]

4
compute/coding-plans/kwai-coding.json
View File

@@ -25,10 +25,6 @@
"serviceType": ["chat"],
"description": "快手 KwaiKAT Coding Plan，编程辅助模型",
"capabilities": ["chat", "code"],
"contextWindow": 128000,
"maxOutputTokens": 8192,
"defaultTemperature": 1,
"defaultTopP": 1,
"source": "preset"
}
]

108
compute/coding-plans/minimax-coding.json
View File

@@ -2,7 +2,7 @@
"id": "provider-minimax-coding-001",
"provider": "minimax",
"brandGroup": "minimax",
"label": "MiniMax Token Plan",
"label": "MiniMax Coding Plan",
"baseUrl": "https://api.minimaxi.com/v1",
"apiFormat": "openai-completions",
"apiKeyRef": "",
@@ -10,109 +10,25 @@
"enabled": false,
"status": "unconfigured",
"accessMode": "coding-plan",
"services": [
"chat",
"reasoning",
"fast"
],
"services": ["chat", "reasoning"],
"codingPlan": {
"quotas": {
"textWindow": "5h",
"standard": {
"starterRequestsPer5h": 1500,
"plusRequestsPer5h": 4500,
"maxRequestsPer5h": 15000
},
"highspeed": {
"plusHighspeedRequestsPer5h": 4500,
"maxHighspeedRequestsPer5h": 15000,
"ultraHighspeedRequestsPer5h": 30000
}
},
"quotas": {},
"usageTracking": {
"method": "rest-api",
"endpoint": "https://www.minimax.io/v1/token_plan/remains",
"consoleUrl": "https://www.minimax.io/user-center/token-plan"
"endpoint": "https://api.minimaxi.com/v1/coding/usage",
"consoleUrl": "https://platform.minimaxi.com/"
}
},
"models": [
{
"modelName": "MiniMax-M2.7",
"displayName": "MiniMax-M2.7 (Token Plan)",
"serviceType": [
"chat"
],
"description": "MiniMax Token Plan 默认文本模型，M2.7，204,800 上下文，约 60 TPS 输出",
"contextWindow": 204800,
"modelName": "MiniMax-M1",
"displayName": "MiniMax-M1 (Coding)",
"serviceType": ["chat"],
"description": "MiniMax Coding Plan，M1 旗舰模型",
"contextWindow": 1000000,
"maxOutputTokens": 131072,
"capabilities": [
"chat",
"reasoning",
"code",
"tool_use",
"long_context"
],
"source": "preset",
"inputPrice": 0.3,
"outputPrice": 1.2,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {
"billing": "token_plan_request_quota",
"cacheReadPrice": 0.06,
"cacheWritePrice": 0.375,
"outputSpeedTps": 60
}
},
{
"modelName": "MiniMax-M2.7-highspeed",
"displayName": "MiniMax-M2.7 高速版 (Token Plan)",
"serviceType": [
"fast"
],
"description": "MiniMax Token Plan 高速订阅支持模型，M2.7-highspeed，效果与 M2.7 一致但输出更快",
"contextWindow": 204800,
"maxOutputTokens": 131072,
"capabilities": [
"chat",
"reasoning",
"code",
"tool_use",
"long_context",
"fast"
],
"source": "preset",
"inputPrice": 0.6,
"outputPrice": 2.4,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {
"billing": "token_plan_highspeed_request_quota",
"cacheReadPrice": 0.06,
"cacheWritePrice": 0.375,
"outputSpeedTps": 100
}
},
{
"modelName": "MiniMax-M2.5",
"displayName": "MiniMax-M2.5 (Coding)",
"serviceType": [
"chat"
],
"description": "MiniMax Token Plan，M2.5 模型，204.8K 上下文",
"contextWindow": 204800,
"maxOutputTokens": 32768,
"capabilities": [
"chat",
"reasoning",
"code",
"long_context"
],
"source": "preset",
"inputPrice": 0.3,
"outputPrice": 1.2,
"defaultTemperature": 1,
"defaultTopP": 0.95
"capabilities": ["chat", "reasoning", "code", "long_context"],
"source": "preset"
}
]
}

75
compute/coding-plans/moonshot-coding.json
View File

@@ -10,77 +10,24 @@
"enabled": false,
"status": "unconfigured",
"accessMode": "coding-plan",
"services": [
"chat",
"reasoning"
],
"services": ["chat", "reasoning"],
"codingPlan": {
"quotas": {
"window": "5h",
"approxApiCallsPerWindow": "300-1200",
"maxConcurrency": 30
},
"quotas": {},
"usageTracking": {
"method": "none",
"consoleUrl": "https://www.kimi.com/code/console"
"consoleUrl": "https://platform.moonshot.cn/"
}
},
"models": [
{
"modelName": "kimi-for-coding",
"displayName": "Kimi for Coding",
"serviceType": [
"chat"
],
"description": "Kimi Code 会员权益模型，官方第三方 Coding Agent 推荐模型 ID，Kimi Code 平台按会员额度计费",
"contextWindow": 262144,
"maxOutputTokens": 32768,
"capabilities": [
"chat",
"reasoning",
"code",
"tool_use",
"agent",
"long_context",
"vision",
"video_understanding"
],
"source": "preset",
"defaultTemperature": null,
"defaultTopP": null,
"extra": {
"billing": "kimi_membership_quota",
"baseUrlNote": "Kimi Code API Key must use https://api.kimi.com/coding/v1",
"recommendedFor": [
"Kimi Code CLI",
"Claude Code",
"Roo Code"
]
}
},
{
"modelName": "kimi-k2.5",
"displayName": "Kimi K2.5 (Coding)",
"serviceType": [
"chat"
],
"description": "Kimi Code 套餐，Kimi K2.5 模型，256K 上下文",
"contextWindow": 256000,
"maxOutputTokens": 32768,
"capabilities": [
"chat",
"reasoning",
"code",
"tool_use",
"agent",
"long_context",
"vision"
],
"source": "preset",
"inputPrice": 0.58,
"outputPrice": 3.02,
"defaultTemperature": 1,
"defaultTopP": 0.95
"modelName": "kimi-k2",
"displayName": "Kimi K2 (Coding)",
"serviceType": ["chat"],
"description": "Kimi Code 套餐，万亿参数 MoE 模型，128K 上下文",
"contextWindow": 128000,
"maxOutputTokens": 8192,
"capabilities": ["chat", "reasoning", "code", "tool_use", "agent", "long_context", "vision"],
"source": "preset"
}
]
}

4
compute/coding-plans/moorethread-coding.json
View File

@@ -25,10 +25,6 @@
"serviceType": ["chat"],
"description": "摩尔线程 Coding Plan，国产 GPU 推理",
"capabilities": ["chat", "code"],
"contextWindow": 128000,
"maxOutputTokens": 8192,
"defaultTemperature": null,
"defaultTopP": null,
"source": "preset"
}
]

4
compute/coding-plans/volcengine-coding.json
View File

@@ -31,10 +31,6 @@
"serviceType": ["chat"],
"description": "自动路由最优编程模型（Doubao-Seed-Code / GLM-4.7 / DeepSeek-V3.2 / Kimi-K2 等）",
"capabilities": ["code", "reasoning", "chat"],
"contextWindow": 256000,
"maxOutputTokens": 32000,
"defaultTemperature": null,
"defaultTopP": null,
"source": "preset"
}
]

117
compute/coding-plans/zhipu-coding.json
View File

@@ -10,10 +10,7 @@
"enabled": false,
"status": "unconfigured",
"accessMode": "coding-plan",
"services": [
"chat",
"reasoning"
],
"services": ["chat", "reasoning"],
"codingPlan": {
"quotas": {},
"usageTracking": {
@@ -22,119 +19,25 @@
}
},
"models": [
{
"modelName": "glm-5.1",
"displayName": "GLM-5.1 (Coding)",
"serviceType": [
"chat"
],
"description": "GLM Coding Plan 全量套餐支持的 GLM-5.1，面向复杂推理和大型工程任务，204,800 上下文，131,072 最大输出",
"contextWindow": 204800,
"maxOutputTokens": 131072,
"capabilities": [
"chat",
"reasoning",
"code",
"deep_thinking",
"long_context",
"tool_use",
"agent",
"math",
"multilingual"
],
"source": "preset",
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {
"quotaMultiplier": {
"peakHoursUtc8": "14:00-18:00",
"peak": 3,
"offPeak": 2,
"promotionalOffPeak": 1,
"promotionalOffPeakEndsAt": "2026-04-30"
}
}
},
{
"modelName": "glm-5-turbo",
"displayName": "GLM-5-Turbo (Coding)",
"serviceType": [
"chat"
],
"description": "GLM Coding Plan 支持的 GLM-5-Turbo，面向 OpenClaw 等长链路任务优化，204,800 上下文，131,072 最大输出",
"contextWindow": 204800,
"maxOutputTokens": 131072,
"capabilities": [
"chat",
"reasoning",
"code",
"deep_thinking",
"long_context",
"tool_use",
"agent"
],
"source": "preset",
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {
"quotaMultiplier": {
"peakHoursUtc8": "14:00-18:00",
"peak": 3,
"offPeak": 2,
"promotionalOffPeak": 1,
"promotionalOffPeakEndsAt": "2026-04-30"
}
}
},
{
"modelName": "glm-5",
"displayName": "GLM-5 (Coding)",
"serviceType": [
"chat"
],
"description": "智谱 GLM-5 Coding Plan 版本，202,752 token 上下文，编程能力登顶",
"contextWindow": 200000,
"serviceType": ["chat"],
"description": "智谱 GLM-5 Coding Plan 版本，745B MoE，编程能力登顶",
"contextWindow": 192000,
"maxOutputTokens": 128000,
"capabilities": [
"chat",
"reasoning",
"code",
"deep_thinking",
"long_context",
"tool_use",
"agent",
"math",
"multilingual"
],
"source": "preset",
"inputPrice": 0.58,
"outputPrice": 2.6,
"defaultTemperature": 1,
"defaultTopP": 0.95
"capabilities": ["chat", "reasoning", "code", "deep_thinking", "long_context", "tool_use", "agent", "math", "multilingual"],
"source": "preset"
},
{
"modelName": "glm-4.7",
"displayName": "GLM-4.7 (Coding)",
"serviceType": [
"chat"
],
"description": "智谱 GLM-4.7 Coding Plan 版本，202,752 token 上下文",
"serviceType": ["chat"],
"description": "智谱 GLM-4.7 Coding Plan 版本，200K 上下文",
"contextWindow": 200000,
"maxOutputTokens": 128000,
"capabilities": [
"chat",
"reasoning",
"code",
"deep_thinking",
"long_context",
"tool_use",
"multilingual"
],
"source": "preset",
"inputPrice": 0.2911,
"outputPrice": 1.1645,
"defaultTemperature": 1,
"defaultTopP": 0.95
"capabilities": ["chat", "reasoning", "code", "deep_thinking", "long_context", "tool_use", "multilingual"],
"source": "preset"
}
]
}

119
compute/providers/anthropic.json
View File

@@ -15,14 +15,12 @@
],
"models": [
{
"modelName": "claude-opus-4-7",
"displayName": "Claude Opus 4.7",
"serviceType": [
"chat"
],
"description": "Anthropic 当前最强通用模型，适合复杂推理和智能体编码任务",
"contextWindow": 1000000,
"maxOutputTokens": 128000,
"modelName": "claude-opus-4-6",
"displayName": "Claude Opus 4.6",
"serviceType": ["chat"],
"description": "Anthropic 最新旗舰 Opus 模型，最强能力",
"contextWindow": 200000,
"maxOutputTokens": 32768,
"capabilities": [
"chat",
"reasoning",
@@ -33,54 +31,32 @@
"inputPrice": 5,
"outputPrice": 25,
"defaultTemperature": 1,
"defaultTopP": null,
"extra": {
"cachePricing": {
"write5m": 6.25,
"write1h": 10,
"read": 0.5
},
"pricingNotes": "Prices are per 1M tokens. Opus 4.7 includes the full 1M context window at standard pricing."
}
"extra": {}
},
{
"modelName": "claude-sonnet-4-6",
"displayName": "Claude Sonnet 4.6",
"serviceType": [
"chat",
"computer_use"
],
"description": "Anthropic 高智能高速度模型，适合编码、工具使用和智能体任务",
"contextWindow": 1000000,
"maxOutputTokens": 64000,
"modelName": "claude-opus-4-5",
"displayName": "Claude Opus 4.5",
"serviceType": ["chat"],
"description": "Anthropic Opus 4.5，顶级推理能力",
"contextWindow": 200000,
"maxOutputTokens": 32768,
"capabilities": [
"chat",
"reasoning",
"code",
"vision",
"tool_use",
"computer_use"
"tool_use"
],
"inputPrice": 3,
"outputPrice": 15,
"inputPrice": 5,
"outputPrice": 25,
"defaultTemperature": 1,
"defaultTopP": null,
"extra": {
"cachePricing": {
"write5m": 3.75,
"write1h": 6,
"read": 0.3
},
"pricingNotes": "Prices are per 1M tokens. Sonnet 4.6 includes the full 1M context window at standard pricing."
}
"extra": {}
},
{
"modelName": "claude-haiku-4-5",
"displayName": "Claude Haiku 4.5",
"serviceType": [
"chat"
],
"description": "Anthropic 最快模型，具备接近前沿模型的智能水平",
"modelName": "claude-sonnet-4-5",
"displayName": "Claude Sonnet 4.5",
"serviceType": ["chat"],
"description": "Anthropic Sonnet 4.5，高性能性价比",
"contextWindow": 200000,
"maxOutputTokens": 64000,
"capabilities": [
@@ -90,19 +66,48 @@
"vision",
"tool_use"
],
"inputPrice": 1,
"outputPrice": 5,
"inputPrice": 3,
"outputPrice": 15,
"defaultTemperature": 1,
"defaultTopP": null,
"extra": {
"modelId": "claude-haiku-4-5-20251001",
"cachePricing": {
"write5m": 1.25,
"write1h": 2,
"read": 0.1
},
"pricingNotes": "Prices are per 1M tokens."
}
"extra": {}
},
{
"modelName": "claude-sonnet-4",
"displayName": "Claude Sonnet 4",
"serviceType": ["chat"],
"description": "Anthropic Sonnet 4 旗舰模型",
"contextWindow": 200000,
"maxOutputTokens": 64000,
"capabilities": [
"chat",
"reasoning",
"code",
"vision",
"tool_use"
],
"inputPrice": 3,
"outputPrice": 15,
"defaultTemperature": 1,
"extra": {}
},
{
"modelName": "claude-sonnet-4-5-computer",
"displayName": "Claude Sonnet 4.5 Computer Use",
"serviceType": ["computer_use"],
"description": "Anthropic 最佳 Computer Use 模型",
"contextWindow": 200000,
"maxOutputTokens": 64000,
"capabilities": [
"chat",
"vision",
"computer_use",
"tool_use",
"reasoning"
],
"inputPrice": 3,
"outputPrice": 15,
"defaultTemperature": 1,
"extra": {}
}
]
}

32
compute/providers/baichuan.json
View File

@@ -18,8 +18,8 @@
"displayName": "百川 M3-Plus",
"serviceType": ["chat"],
"description": "百川智能M3-Plus医疗增强模型，235B参数，幻觉率仅2.6%，API价格较前代下降70%",
"contextWindow": 32000,
"maxOutputTokens": 32000,
"contextWindow": 192000,
"maxOutputTokens": 8192,
"capabilities": [
"chat",
"reasoning",
@@ -31,8 +31,8 @@
],
"inputPrice": 5,
"outputPrice": 9,
"defaultTemperature": null,
"defaultTopP": null,
"defaultTemperature": 0.7,
"defaultTopP": 0.9,
"extra": {}
},
{
@@ -40,8 +40,8 @@
"displayName": "百川 M3",
"serviceType": ["chat"],
"description": "百川智能M3新一代开源医疗增强模型，235B参数",
"contextWindow": 32000,
"maxOutputTokens": 32000,
"contextWindow": 192000,
"maxOutputTokens": 8192,
"capabilities": [
"chat",
"reasoning",
@@ -52,8 +52,8 @@
],
"inputPrice": 10,
"outputPrice": 30,
"defaultTemperature": null,
"defaultTopP": null,
"defaultTemperature": 0.7,
"defaultTopP": 0.9,
"extra": {}
},
{
@@ -61,8 +61,8 @@
"displayName": "百川 M2-Plus",
"serviceType": ["chat"],
"description": "百川智能M2-Plus旗舰模型，最强推理和创作能力",
"contextWindow": 32000,
"maxOutputTokens": 32000,
"contextWindow": 128000,
"maxOutputTokens": 8192,
"capabilities": [
"chat",
"reasoning",
@@ -72,8 +72,8 @@
],
"inputPrice": 10,
"outputPrice": 30,
"defaultTemperature": null,
"defaultTopP": null,
"defaultTemperature": 0.7,
"defaultTopP": 0.9,
"extra": {}
},
{
@@ -81,8 +81,8 @@
"displayName": "百川 M2",
"serviceType": ["chat"],
"description": "百川智能M2模型，高性能通用模型",
"contextWindow": 32000,
"maxOutputTokens": 32000,
"contextWindow": 128000,
"maxOutputTokens": 8192,
"capabilities": [
"chat",
"reasoning",
@@ -91,8 +91,8 @@
],
"inputPrice": 2,
"outputPrice": 20,
"defaultTemperature": null,
"defaultTopP": null,
"defaultTemperature": 0.7,
"defaultTopP": 0.9,
"extra": {}
}
]

134
compute/providers/baidu.json
View File

@@ -9,20 +9,14 @@
"enabled": false,
"status": "unconfigured",
"priceCurrency": "CNY",
"services": [
"chat",
"reasoning"
],
"services": ["chat", "reasoning"],
"models": [
{
"modelName": "ernie-5.0-thinking-latest",
"displayName": "文心 ERNIE 5.0",
"serviceType": [
"chat",
"reasoning"
],
"serviceType": ["chat", "reasoning"],
"description": "百度最新旗舰模型，支持深度推理、多模态和工具调用，119K输入/64K输出",
"contextWindow": 128000,
"contextWindow": 131072,
"maxOutputTokens": 65536,
"capabilities": [
"chat",
@@ -31,69 +25,19 @@
"multimodal",
"tool_use"
],
"inputPrice": 6,
"outputPrice": 24,
"defaultTemperature": null,
"defaultTopP": null,
"extra": {
"pricingTiers": [
{
"maxInputTokens": 32768,
"inputPrice": 6,
"outputPrice": 24
},
{
"maxInputTokens": 131072,
"inputPrice": 10,
"outputPrice": 40
}
],
"thinkingMaxTokens": 60000
}
"inputPrice": 8,
"outputPrice": 32,
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {}
},
{
"modelName": "ernie-5.0",
"displayName": "文心 ERNIE 5.0 非思考",
"serviceType": [
"chat"
],
"description": "百度 ERNIE 5.0 旗舰模型，128K 上下文，119K 最大输入，最大输出 65,536 token",
"contextWindow": 131072,
"maxOutputTokens": 65536,
"capabilities": [
"chat",
"vision",
"multimodal",
"tool_use"
],
"inputPrice": 6,
"outputPrice": 24,
"defaultTemperature": null,
"defaultTopP": null,
"extra": {
"pricingTiers": [
{
"maxInputTokens": 32768,
"inputPrice": 6,
"outputPrice": 24
},
{
"maxInputTokens": 131072,
"inputPrice": 10,
"outputPrice": 40
}
]
}
},
{
"modelName": "ernie-4.5-turbo-128k",
"modelName": "ernie-4.5-turbo-128k-latest",
"displayName": "文心 ERNIE 4.5 Turbo",
"serviceType": [
"chat"
],
"description": "百度高性价比长上下文模型，128K 窗口，最大输出 12,288 token",
"serviceType": ["chat"],
"description": "百度高性价比长上下文模型，128K 窗口，适合日常对话和文档处理",
"contextWindow": 131072,
"maxOutputTokens": 12288,
"maxOutputTokens": 8192,
"capabilities": [
"chat",
"code",
@@ -103,59 +47,9 @@
],
"inputPrice": 0.8,
"outputPrice": 3.2,
"defaultTemperature": 0.8,
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {
"cacheHitPrice": 0.2
}
},
{
"modelName": "ernie-4.5-turbo-20260402",
"displayName": "文心 ERNIE 4.5 Turbo 20260402",
"serviceType": [
"chat"
],
"description": "百度 ERNIE 4.5 Turbo 20260402 快照，128K 上下文，最大输出 12,288 token",
"contextWindow": 131072,
"maxOutputTokens": 12288,
"capabilities": [
"chat",
"code",
"vision",
"long_context",
"fast"
],
"inputPrice": 0.8,
"outputPrice": 3.2,
"defaultTemperature": 0.8,
"defaultTopP": 1,
"extra": {
"cacheHitPrice": 0.2
}
},
{
"modelName": "ernie-x1.1",
"displayName": "文心 ERNIE X1.1",
"serviceType": [
"reasoning"
],
"description": "百度 ERNIE X1.1 深度思考模型，64K 上下文，最大输出 65,536 token",
"contextWindow": 65536,
"maxOutputTokens": 65536,
"capabilities": [
"chat",
"reasoning",
"deep_thinking",
"math",
"code"
],
"inputPrice": 1,
"outputPrice": 4,
"defaultTemperature": null,
"defaultTopP": null,
"extra": {
"thinkingMaxTokens": 65536
}
"extra": {}
}
]
}

58
compute/providers/cohere.json
View File

@@ -15,9 +15,9 @@
"modelName": "command-a-03-2025",
"displayName": "Command A",
"serviceType": ["chat"],
"description": "Cohere 旗舰对话模型，256K 上下文，擅长 RAG 和工具调用",
"contextWindow": 256000,
"maxOutputTokens": 8000,
"description": "Cohere 旗舰对话模型，262K 上下文，擅长 RAG 和工具调用",
"contextWindow": 262144,
"maxOutputTokens": 8192,
"capabilities": [
"chat",
"reasoning",
@@ -29,40 +29,16 @@
],
"inputPrice": 2.50,
"outputPrice": 10.00,
"defaultTemperature": 0.3,
"defaultTopP": 0.75,
"extra": {
"pricingNotes": "Prices are per 1M tokens."
}
},
{
"modelName": "command-r7b-12-2024",
"displayName": "Command R7B",
"serviceType": ["fast"],
"description": "Cohere 小型高速对话模型，适合高吞吐、低延迟场景",
"contextWindow": 128000,
"maxOutputTokens": 4000,
"capabilities": [
"chat",
"reasoning",
"tool_use",
"rag",
"fast"
],
"inputPrice": 0.0375,
"outputPrice": 0.15,
"defaultTemperature": 0.3,
"defaultTopP": 0.75,
"extra": {
"pricingNotes": "Prices are per 1M tokens."
}
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {}
},
{
"modelName": "embed-v4.0",
"displayName": "Embed V4",
"serviceType": ["embedding"],
"description": "Cohere 最新 Embedding 模型，128K 上下文，多语言支持",
"contextWindow": 128000,
"description": "Cohere 最新 Embedding 模型，131K 上下文，多语言支持",
"contextWindow": 131072,
"maxOutputTokens": 0,
"capabilities": [
"text_embedding",
@@ -70,20 +46,14 @@
],
"inputPrice": 0.12,
"outputPrice": 0,
"defaultTemperature": null,
"defaultTopP": null,
"extra": {
"dimensions": [256, 512, 1024, 1536],
"defaultDimension": 1536,
"pricingNotes": "Embedding models are priced by embedded tokens; price is per 1M tokens."
}
"extra": {}
},
{
"modelName": "rerank-v3.5",
"displayName": "Rerank V3.5",
"serviceType": ["rerank"],
"description": "Cohere 语义重排序模型，用于检索结果精排，默认单文档截断 4096 token",
"contextWindow": 4096,
"description": "Cohere 语义重排序模型，用于检索结果精排",
"contextWindow": 0,
"maxOutputTokens": 0,
"capabilities": [
"rerank",
@@ -91,11 +61,7 @@
],
"inputPrice": 2.00,
"outputPrice": 0,
"defaultTemperature": null,
"defaultTopP": null,
"extra": {
"pricingNotes": "Cohere Rerank is priced by search units, not input/output tokens; one search unit is one query with up to 100 documents. The token price fields are retained for schema compatibility."
}
"extra": {}
}
]
}

286
compute/providers/dashscope.json
View File

@@ -23,97 +23,13 @@
"translation"
],
"models": [
{
"modelName": "qwen3.6-plus",
"displayName": "阿里云 Qwen3.6-Plus",
"serviceType": [
"chat",
"vision"
],
"description": "通义千问3.6 Plus，支持文本/图像/视频输入，100万上下文，支持内置工具和 Function Calling",
"contextWindow": 1000000,
"maxOutputTokens": 65536,
"capabilities": [
"chat",
"reasoning",
"code",
"multilingual",
"long_context",
"tool_use",
"agent",
"vision"
],
"inputPrice": 2,
"outputPrice": 12,
"defaultTemperature": 0.6,
"defaultTopP": 0.95,
"extra": {
"pricingTiers": [
{
"maxInputTokens": 256000,
"inputPrice": 2,
"outputPrice": 12,
"thinkingOutputPrice": 12
},
{
"maxInputTokens": 1000000,
"inputPrice": 8,
"outputPrice": 48,
"thinkingOutputPrice": 48
}
],
"thinkingMaxTokens": 81920
}
},
{
"modelName": "qwen3.6-flash",
"displayName": "阿里云 Qwen3.6-Flash",
"serviceType": [
"chat",
"vision"
],
"description": "通义千问3.6 Flash，高性价比多模态模型，100万上下文，支持内置工具和 Function Calling",
"contextWindow": 1000000,
"maxOutputTokens": 65536,
"capabilities": [
"chat",
"reasoning",
"code",
"multilingual",
"long_context",
"tool_use",
"vision",
"fast"
],
"inputPrice": 1.2,
"outputPrice": 7.2,
"defaultTemperature": 0.6,
"defaultTopP": 0.95,
"extra": {
"pricingTiers": [
{
"maxInputTokens": 256000,
"inputPrice": 1.2,
"outputPrice": 7.2
},
{
"maxInputTokens": 1000000,
"inputPrice": 4.8,
"outputPrice": 28.8
}
],
"thinkingMaxTokens": 131072
}
},
{
"modelName": "qwen3.5-plus",
"displayName": "阿里云 Qwen3.5-Plus",
"serviceType": [
"chat"
],
"description": "通义千问3.5旗舰模型，支持文本/图像/视频输入，100万上下文，默认开启思考模式",
"contextWindow": 1000000,
"maxOutputTokens": 65536,
"serviceType": ["chat"],
"description": "通义千问3.5旗舰模型，支持文本/图像/视频输入，效果媲美Qwen3 Max，性价比更优",
"contextWindow": 262144,
"maxOutputTokens": 16384,
"capabilities": [
"chat",
"reasoning",
@@ -126,37 +42,16 @@
],
"inputPrice": 0.8,
"outputPrice": 4.8,
"defaultTemperature": 0.6,
"defaultTopP": 0.95,
"extra": {
"pricingTiers": [
{
"maxInputTokens": 128000,
"inputPrice": 0.8,
"outputPrice": 4.8
},
{
"maxInputTokens": 256000,
"inputPrice": 2,
"outputPrice": 12
},
{
"maxInputTokens": 1000000,
"inputPrice": 4,
"outputPrice": 24
}
],
"thinkingMaxTokens": 81920
}
"defaultTemperature": 0.7,
"defaultTopP": 0.8,
"extra": {}
},
{
"modelName": "qwen-max",
"displayName": "阿里云 qwen-max",
"serviceType": [
"chat"
],
"description": "通义千问Max，千亿参数旗舰模型，32K 上下文",
"contextWindow": 32768,
"serviceType": ["chat"],
"description": "通义千问Max，千亿参数旗舰模型，最强性能",
"contextWindow": 131072,
"maxOutputTokens": 8192,
"capabilities": [
"chat",
@@ -174,12 +69,10 @@
{
"modelName": "qwen-plus",
"displayName": "阿里云 qwen-plus",
"serviceType": [
"chat"
],
"description": "通义千问Plus，支持思考模式，100万上下文",
"contextWindow": 1000000,
"maxOutputTokens": 32768,
"serviceType": ["chat"],
"description": "通义千问Plus，支持思考模式，128K 上下文",
"contextWindow": 131072,
"maxOutputTokens": 8192,
"capabilities": [
"chat",
"reasoning",
@@ -188,66 +81,36 @@
],
"inputPrice": 0.8,
"outputPrice": 2,
"defaultTemperature": null,
"defaultTopP": null,
"extra": {
"pricingTiers": [
{
"maxInputTokens": 128000,
"inputPrice": 0.8,
"outputPrice": 2,
"thinkingOutputPrice": 8
},
{
"maxInputTokens": 256000,
"inputPrice": 2.4,
"outputPrice": 20,
"thinkingOutputPrice": 24
},
{
"maxInputTokens": 1000000,
"inputPrice": 4.8,
"outputPrice": 48,
"thinkingOutputPrice": 64
}
],
"thinkingMaxTokens": 81920
}
"defaultTemperature": 0.6,
"defaultTopP": 0.8,
"extra": {}
},
{
"modelName": "qwen-turbo",
"displayName": "阿里云 qwen-turbo",
"serviceType": [
"chat"
],
"description": "通义千问Turbo，非思考模式支持100万上下文，思考模式支持131K上下文",
"serviceType": ["chat"],
"description": "通义千问Turbo，支持100万上下文，高性价比",
"contextWindow": 1000000,
"maxOutputTokens": 16384,
"maxOutputTokens": 8192,
"capabilities": [
"chat",
"code",
"long_context",
"fast"
],
"inputPrice": 0.367,
"outputPrice": 1.468,
"defaultTemperature": null,
"defaultTopP": null,
"extra": {
"thinkingContextWindow": 131072,
"thinkingOutputPrice": 3.67,
"thinkingMaxTokens": 38912
}
"inputPrice": 0.3,
"outputPrice": 0.6,
"defaultTemperature": 0.7,
"defaultTopP": 0.8,
"extra": {}
},
{
"modelName": "qwen-long",
"displayName": "阿里云 qwen-long",
"serviceType": [
"chat"
],
"serviceType": ["chat"],
"description": "通义千问Long，支持1000万上下文，适合超长文档处理，降价97%",
"contextWindow": 10000000,
"maxOutputTokens": 32768,
"maxOutputTokens": 8192,
"capabilities": [
"chat",
"ultra_long_context",
@@ -255,16 +118,14 @@
],
"inputPrice": 0.5,
"outputPrice": 2,
"defaultTemperature": 1,
"defaultTemperature": 0.7,
"defaultTopP": 0.8,
"extra": {}
},
{
"modelName": "qwen3-max",
"displayName": "阿里云 Qwen3-Max",
"serviceType": [
"chat"
],
"serviceType": ["chat"],
"description": "通义千问3代旗舰模型，262K 上下文，支持思考模式",
"contextWindow": 262144,
"maxOutputTokens": 65536,
@@ -280,19 +141,17 @@
],
"inputPrice": 2.5,
"outputPrice": 10,
"defaultTemperature": null,
"defaultTopP": null,
"defaultTemperature": 0.7,
"defaultTopP": 0.8,
"extra": {}
},
{
"modelName": "qwen3-vl-plus",
"displayName": "阿里云 Qwen3-VL-Plus",
"serviceType": [
"vision"
],
"serviceType": ["vision"],
"description": "通义千问3代视觉模型Plus版，支持图像理解",
"contextWindow": 262144,
"maxOutputTokens": 32768,
"contextWindow": 131072,
"maxOutputTokens": 8192,
"capabilities": [
"chat",
"vision",
@@ -302,19 +161,15 @@
],
"inputPrice": 1.5,
"outputPrice": 6,
"defaultTemperature": 0.7,
"defaultTopP": 0.8,
"extra": {}
},
{
"modelName": "qwen3-vl-flash",
"displayName": "阿里云 Qwen3-VL-Flash",
"serviceType": [
"vision"
],
"serviceType": ["vision"],
"description": "通义千问3代视觉模型Flash版，高性价比",
"contextWindow": 262144,
"maxOutputTokens": 32768,
"contextWindow": 131072,
"maxOutputTokens": 8192,
"capabilities": [
"chat",
"vision",
@@ -323,78 +178,57 @@
],
"inputPrice": 0.8,
"outputPrice": 3,
"defaultTemperature": 0.7,
"defaultTopP": 0.8,
"extra": {}
},
{
"modelName": "text-embedding-v3",
"displayName": "text-embedding-v3",
"serviceType": [
"embedding"
],
"serviceType": ["embedding"],
"description": "阿里云文本向量化",
"contextWindow": 8192,
"capabilities": [
"text_embedding",
"semantic_search",
"chinese_optimized"
],
"inputPrice": 0.7,
"defaultTemperature": null,
"defaultTopP": null,
"extra": {}
},
{
"modelName": "text-embedding-v4",
"displayName": "阿里云 text-embedding-v4",
"serviceType": [
"embedding"
],
"serviceType": ["embedding"],
"description": "阿里云百炼文本向量模型v4，支持多维度与多语种",
"contextWindow": 8192,
"capabilities": [
"text_embedding",
"semantic_search",
"rag"
],
"inputPrice": 0.5,
"defaultTemperature": null,
"defaultTopP": null,
"extra": {}
},
{
"modelName": "qwen3-rerank",
"displayName": "Qwen3-Rerank",
"serviceType": [
"rerank"
],
"serviceType": ["rerank"],
"description": "检索重排序，支持 100+ 语种",
"contextWindow": 120000,
"capabilities": [
"rerank",
"semantic_reranking",
"multilingual"
],
"inputPrice": 1,
"defaultTemperature": null,
"defaultTopP": null,
"extra": {}
},
{
"modelName": "cosyvoice-v2",
"displayName": "CosyVoice V2",
"serviceType": [
"tts"
],
"serviceType": ["tts"],
"description": "通义语音合成，支持中英文",
"capabilities": [
"tts",
"multilingual",
"chinese_optimized"
],
"defaultTemperature": null,
"defaultTopP": null,
"extra": {
"voices": [
"longxiaochun",
@@ -412,16 +246,12 @@
{
"modelName": "paraformer-v2",
"displayName": "Paraformer V2",
"serviceType": [
"asr"
],
"serviceType": ["asr"],
"description": "通义语音识别",
"capabilities": [
"asr",
"chinese_optimized"
],
"defaultTemperature": null,
"defaultTopP": null,
"extra": {
"maxAudioLength": 3600,
"supportedInputFormats": [
@@ -436,16 +266,12 @@
{
"modelName": "wanx-v2",
"displayName": "通义万相 V2",
"serviceType": [
"image_gen"
],
"serviceType": ["image_gen"],
"description": "通义万相文生图",
"capabilities": [
"image_generation",
"chinese_optimized"
],
"defaultTemperature": null,
"defaultTopP": null,
"extra": {
"supportedImageSizes": [
"1024x1024",
@@ -463,16 +289,12 @@
{
"modelName": "wanx-video",
"displayName": "通义万相视频",
"serviceType": [
"video_gen"
],
"serviceType": ["video_gen"],
"description": "通义万相文生视频",
"capabilities": [
"video_generation",
"chinese_optimized"
],
"defaultTemperature": null,
"defaultTopP": null,
"extra": {
"maxVideoDuration": 10,
"supportedResolutions": [
@@ -484,17 +306,13 @@
{
"modelName": "cosyvoice-clone",
"displayName": "CosyVoice 声音复刻",
"serviceType": [
"voice_clone"
],
"serviceType": ["voice_clone"],
"description": "通义声音复刻，少量样本即可克隆",
"capabilities": [
"voice_clone",
"chinese_optimized",
"few_shot"
],
"defaultTemperature": null,
"defaultTopP": null,
"extra": {
"maxCloneSamples": 5,
"minCloneDuration": 3
@@ -503,12 +321,9 @@
{
"modelName": "qwen-omni-turbo",
"displayName": "Qwen-Omni-Turbo",
"serviceType": [
"omni"
],
"serviceType": ["omni"],
"description": "通义全能模型，支持音视频文本多模态",
"contextWindow": 32768,
"maxOutputTokens": 2048,
"capabilities": [
"chat",
"vision",
@@ -517,26 +332,19 @@
"multimodal",
"tool_use"
],
"defaultTemperature": null,
"defaultTopP": 0.01,
"extra": {}
},
{
"modelName": "qwen3-max-trans",
"displayName": "Qwen3-Max（翻译）",
"serviceType": [
"translation"
],
"serviceType": ["translation"],
"description": "通义千问翻译，中英日韩等多语种互译",
"contextWindow": 131072,
"maxOutputTokens": 8192,
"capabilities": [
"translation",
"multilingual",
"chinese_optimized"
],
"defaultTemperature": null,
"defaultTopP": null,
"extra": {}
}
]

34
compute/providers/deepseek.json
View File

@@ -17,12 +17,10 @@
{
"modelName": "deepseek-chat",
"displayName": "DeepSeek V3.2",
"serviceType": [
"chat"
],
"description": "高性价比通用对话模型，128K 上下文",
"contextWindow": 128000,
"maxOutputTokens": 8192,
"serviceType": ["chat"],
"description": "高性价比通用对话模型，64K 上下文",
"contextWindow": 64000,
"maxOutputTokens": 8000,
"capabilities": [
"chat",
"code",
@@ -30,22 +28,18 @@
"multilingual"
],
"inputPrice": 2,
"outputPrice": 3,
"outputPrice": 8,
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {
"cacheHitPrice": 0.2
}
"extra": {}
},
{
"modelName": "deepseek-reasoner",
"displayName": "DeepSeek R1",
"serviceType": [
"reasoning"
],
"description": "DeepSeek V3.2 思考模式，128K 上下文，思维链默认 32K，最大输出 64K",
"contextWindow": 128000,
"maxOutputTokens": 65536,
"serviceType": ["reasoning"],
"description": "DeepSeek R1 推理模型，64K 上下文，深度推理能力，思维链最长 32K",
"contextWindow": 64000,
"maxOutputTokens": 8192,
"capabilities": [
"chat",
"reasoning",
@@ -55,11 +49,9 @@
],
"inputPrice": 4,
"outputPrice": 16,
"defaultTemperature": null,
"defaultTopP": null,
"extra": {
"cacheHitPrice": 0.2
}
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {}
}
]
}

101
compute/providers/google.json
View File

@@ -15,98 +15,12 @@
"embedding"
],
"models": [
{
"modelName": "gemini-3.1-pro-preview",
"displayName": "Gemini 3.1 Pro Preview",
"serviceType": ["chat"],
"description": "Google Gemini 3.1 Pro Preview，面向复杂推理、编码和智能体工作流",
"contextWindow": 1048576,
"maxOutputTokens": 65536,
"capabilities": [
"chat",
"reasoning",
"code",
"vision",
"ultra_long_context",
"tool_use"
],
"inputPrice": 2.00,
"outputPrice": 12.00,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {
"pricingNotes": "Standard paid tier text/image/video price for prompts <= 200K tokens; prompts > 200K are $4 input and $18 output per 1M tokens.",
"cachePricing": {
"inputCacheRead": 0.20,
"inputCacheReadOver200k": 0.40,
"storagePerMillionTokensPerHour": 4.50
}
}
},
{
"modelName": "gemini-3-flash-preview",
"displayName": "Gemini 3 Flash Preview",
"serviceType": ["chat"],
"description": "Google Gemini 3 Flash Preview，高速多模态推理模型",
"contextWindow": 1048576,
"maxOutputTokens": 65536,
"capabilities": [
"chat",
"reasoning",
"code",
"vision",
"ultra_long_context",
"tool_use",
"fast"
],
"inputPrice": 0.50,
"outputPrice": 3.00,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {
"pricingNotes": "Standard paid tier text/image/video price; audio input is $1.00 per 1M tokens.",
"cachePricing": {
"inputCacheRead": 0.05,
"audioInputCacheRead": 0.10,
"storagePerMillionTokensPerHour": 1.00
}
}
},
{
"modelName": "gemini-3.1-flash-lite-preview",
"displayName": "Gemini 3.1 Flash-Lite Preview",
"serviceType": ["fast"],
"description": "Google Gemini 3.1 Flash-Lite Preview，低成本高吞吐多模态模型",
"contextWindow": 1048576,
"maxOutputTokens": 65536,
"capabilities": [
"chat",
"reasoning",
"code",
"vision",
"ultra_long_context",
"tool_use",
"fast"
],
"inputPrice": 0.25,
"outputPrice": 1.50,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {
"pricingNotes": "Standard paid tier text/image/video price; audio input is $0.50 per 1M tokens.",
"cachePricing": {
"inputCacheRead": 0.025,
"audioInputCacheRead": 0.05,
"storagePerMillionTokensPerHour": 1.00
}
}
},
{
"modelName": "gemini-2.5-pro",
"displayName": "Gemini 2.5 Pro",
"serviceType": ["chat"],
"description": "Google 最新旗舰模型，1,048,576 token 上下文，多模态",
"contextWindow": 1048576,
"description": "Google 最新旗舰模型，100万上下文，多模态",
"contextWindow": 1000000,
"maxOutputTokens": 65536,
"capabilities": [
"chat",
@@ -118,7 +32,7 @@
],
"inputPrice": 1.25,
"outputPrice": 10,
"defaultTemperature": null,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {}
},
@@ -126,8 +40,8 @@
"modelName": "gemini-2.5-flash",
"displayName": "Gemini 2.5 Flash",
"serviceType": ["chat"],
"description": "Google 高性价比模型，1,048,576 token 上下文，速度快",
"contextWindow": 1048576,
"description": "Google 高性价比模型，100万上下文，速度快",
"contextWindow": 1000000,
"maxOutputTokens": 65536,
"capabilities": [
"chat",
@@ -140,7 +54,7 @@
],
"inputPrice": 0.30,
"outputPrice": 2.50,
"defaultTemperature": null,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {}
},
@@ -149,14 +63,11 @@
"displayName": "Text Embedding 005",
"serviceType": ["embedding"],
"description": "Google 文本嵌入模型，768维度",
"contextWindow": 2048,
"capabilities": [
"text_embedding",
"semantic_search"
],
"inputPrice": 0.10,
"defaultTemperature": null,
"defaultTopP": null,
"extra": {}
}
]

60
compute/providers/internal-testing.json
View File

@@ -23,7 +23,7 @@
"chat"
],
"description": "MiniMax M2.7 高速版，低延迟吞吐优化，200K 上下文",
"contextWindow": 204800,
"contextWindow": 200000,
"maxOutputTokens": 8192,
"capabilities": [
"chat",
@@ -33,7 +33,6 @@
"inputPrice": 0,
"outputPrice": 0,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {}
},
{
@@ -42,9 +41,9 @@
"serviceType": [
"chat"
],
"description": "智谱 GLM-5.1 新一代旗舰模型，编程与推理能力断档领先，200K 上下文",
"contextWindow": 200000,
"maxOutputTokens": 128000,
"description": "智谱 GLM-5.1 新一代旗舰模型，编程与推理能力断档领先，204K 上下文",
"contextWindow": 204800,
"maxOutputTokens": 131072,
"capabilities": [
"chat",
"reasoning",
@@ -55,8 +54,7 @@
],
"inputPrice": 0,
"outputPrice": 0,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"defaultTemperature": 0.95,
"extra": {}
},
{
@@ -65,9 +63,9 @@
"serviceType": [
"chat"
],
"description": "智谱 GLM-5 大语言模型，200K 上下文",
"contextWindow": 200000,
"maxOutputTokens": 128000,
"description": "智谱 GLM-5 大语言模型",
"contextWindow": 128000,
"maxOutputTokens": 8192,
"capabilities": [
"chat",
"reasoning",
@@ -76,7 +74,6 @@
"inputPrice": 0,
"outputPrice": 0,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {}
},
{
@@ -86,8 +83,8 @@
"chat"
],
"description": "智谱 GLM-5-Turbo 大语言模型，更快的推理速度",
"contextWindow": 200000,
"maxOutputTokens": 128000,
"contextWindow": 128000,
"maxOutputTokens": 8192,
"capabilities": [
"chat",
"reasoning",
@@ -96,7 +93,6 @@
"inputPrice": 0,
"outputPrice": 0,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {}
},
{
@@ -116,7 +112,6 @@
"inputPrice": 0,
"outputPrice": 0,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {}
},
{
@@ -127,7 +122,7 @@
],
"description": "月之暗面 Kimi K2.6 Code Preview，万亿参数 MoE 架构（32B 激活），256K 上下文，代码与智能体能力增强",
"contextWindow": 256000,
"maxOutputTokens": 32768,
"maxOutputTokens": 16384,
"capabilities": [
"chat",
"reasoning",
@@ -139,8 +134,7 @@
],
"inputPrice": 0,
"outputPrice": 0,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"defaultTemperature": 0.7,
"extra": {}
},
{
@@ -150,8 +144,8 @@
"chat"
],
"description": "月之暗面 Kimi-2.5 大语言模型",
"contextWindow": 256000,
"maxOutputTokens": 32768,
"contextWindow": 128000,
"maxOutputTokens": 8192,
"capabilities": [
"chat",
"reasoning",
@@ -164,7 +158,6 @@
"inputPrice": 0,
"outputPrice": 0,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {}
},
{
@@ -174,7 +167,7 @@
"chat"
],
"description": "MiniMax M2.5 大语言模型",
"contextWindow": 204800,
"contextWindow": 128000,
"maxOutputTokens": 8192,
"capabilities": [
"chat",
@@ -186,7 +179,6 @@
"inputPrice": 0,
"outputPrice": 0,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {}
},
{
@@ -210,7 +202,6 @@
"inputPrice": 0,
"outputPrice": 0,
"defaultTemperature": 0.7,
"defaultTopP": 0.8,
"extra": {}
},
{
@@ -220,8 +211,8 @@
"chat"
],
"description": "阿里通义千问 Qwen3.5 Plus",
"contextWindow": 1000000,
"maxOutputTokens": 65536,
"contextWindow": 128000,
"maxOutputTokens": 8192,
"capabilities": [
"chat",
"reasoning",
@@ -233,8 +224,7 @@
],
"inputPrice": 0,
"outputPrice": 0,
"defaultTemperature": 0.7,
"defaultTopP": 0.8,
"defaultTemperature": 1,
"extra": {}
},
{
@@ -256,7 +246,6 @@
"inputPrice": 0,
"outputPrice": 0,
"defaultTemperature": 0.7,
"defaultTopP": 0.8,
"extra": {}
},
{
@@ -278,7 +267,6 @@
"inputPrice": 0,
"outputPrice": 0,
"defaultTemperature": 0.7,
"defaultTopP": 0.8,
"extra": {}
},
{
@@ -288,8 +276,8 @@
"chat"
],
"description": "阿里通义千问 Qwen3 Max (2026-01-23)",
"contextWindow": 262144,
"maxOutputTokens": 32768,
"contextWindow": 128000,
"maxOutputTokens": 8192,
"capabilities": [
"chat",
"reasoning",
@@ -301,8 +289,7 @@
],
"inputPrice": 0,
"outputPrice": 0,
"defaultTemperature": 0.7,
"defaultTopP": 0.8,
"defaultTemperature": 1,
"extra": {}
},
{
@@ -312,8 +299,8 @@
"chat"
],
"description": "字节跳动豆包 Seed 2.0 Code Preview (260215)",
"contextWindow": 256000,
"maxOutputTokens": 128000,
"contextWindow": 128000,
"maxOutputTokens": 8192,
"capabilities": [
"chat",
"reasoning",
@@ -322,7 +309,6 @@
"inputPrice": 0,
"outputPrice": 0,
"defaultTemperature": 1,
"defaultTopP": 0.7,
"extra": {}
}
]

8
compute/providers/lingyiwanwu.json
View File

@@ -30,8 +30,8 @@
],
"inputPrice": 0.99,
"outputPrice": 0.99,
"defaultTemperature": null,
"defaultTopP": null,
"defaultTemperature": 0.7,
"defaultTopP": 0.9,
"extra": {}
},
{
@@ -51,8 +51,8 @@
],
"inputPrice": 6,
"outputPrice": 6,
"defaultTemperature": null,
"defaultTopP": null,
"defaultTemperature": 0.7,
"defaultTopP": 0.9,
"extra": {}
}
]

102
compute/providers/minimax.json
View File

@@ -9,19 +9,12 @@
"enabled": false,
"status": "unconfigured",
"priceCurrency": "CNY",
"services": [
"chat",
"fast",
"reasoning"
],
"services": ["chat", "fast", "reasoning"],
"models": [
{
"modelName": "MiniMax-M2.7",
"displayName": "MiniMax M2.7",
"serviceType": [
"chat",
"reasoning"
],
"serviceType": ["chat", "reasoning"],
"description": "MiniMax 新一代旗舰模型，擅长复杂 Agent、软件工程与专业办公任务",
"contextWindow": 204800,
"maxOutputTokens": 131072,
@@ -31,22 +24,14 @@
"code",
"tool_use"
],
"inputPrice": 2.1,
"outputPrice": 8.4,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {
"cacheReadPrice": 0.42,
"cacheWritePrice": 2.625,
"outputSpeedTps": 60
}
"defaultTopP": 1,
"extra": {}
},
{
"modelName": "MiniMax-M2.7-highspeed",
"displayName": "MiniMax M2.7 高速版",
"serviceType": [
"fast"
],
"serviceType": ["fast"],
"description": "MiniMax M2.7 极速版，效果一致，输出更快，适合低延迟场景",
"contextWindow": 204800,
"maxOutputTokens": 131072,
@@ -57,25 +42,16 @@
"tool_use",
"fast"
],
"inputPrice": 4.2,
"outputPrice": 16.8,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {
"cacheReadPrice": 0.42,
"cacheWritePrice": 2.625,
"outputSpeedTps": 100
}
"defaultTopP": 1,
"extra": {}
},
{
"modelName": "MiniMax-M2.5",
"displayName": "MiniMax M2.5",
"serviceType": [
"chat",
"reasoning"
],
"description": "MiniMax M2.5，204.8K 上下文，支持深度推理和工具调用",
"contextWindow": 204800,
"serviceType": ["chat", "reasoning"],
"description": "MiniMax 最新旗舰模型，230B参数(10B激活)MoE架构，百万级上下文，支持深度推理和工具调用",
"contextWindow": 1000000,
"maxOutputTokens": 131072,
"capabilities": [
"chat",
@@ -87,21 +63,15 @@
"inputPrice": 2.1,
"outputPrice": 8.4,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {
"cacheReadPrice": 0.21,
"cacheWritePrice": 2.625,
"outputSpeedTps": 60
}
"defaultTopP": 1,
"extra": {}
},
{
"modelName": "MiniMax-M2.5-highspeed",
"displayName": "MiniMax M2.5 高速版",
"serviceType": [
"fast"
],
"serviceType": ["fast"],
"description": "MiniMax M2.5 高速推理版本(100TPS)，适合低延迟场景",
"contextWindow": 204800,
"contextWindow": 200000,
"maxOutputTokens": 131072,
"capabilities": [
"chat",
@@ -110,22 +80,16 @@
"tool_use",
"fast"
],
"inputPrice": 4.2,
"inputPrice": 2.1,
"outputPrice": 16.8,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {
"cacheReadPrice": 0.21,
"cacheWritePrice": 2.625,
"outputSpeedTps": 100
}
"defaultTopP": 1,
"extra": {}
},
{
"modelName": "MiniMax-M2.1",
"displayName": "MiniMax M2.1",
"serviceType": [
"chat"
],
"serviceType": ["chat"],
"description": "MiniMax 上一代主力模型，200K 上下文，性价比优秀",
"contextWindow": 204800,
"maxOutputTokens": 131072,
@@ -138,19 +102,13 @@
"inputPrice": 2.1,
"outputPrice": 8.4,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {
"cacheReadPrice": 0.21,
"cacheWritePrice": 2.625,
"outputSpeedTps": 60
}
"defaultTopP": 1,
"extra": {}
},
{
"modelName": "MiniMax-M2.1-highspeed",
"displayName": "MiniMax M2.1 高速版",
"serviceType": [
"fast"
],
"serviceType": ["fast"],
"description": "MiniMax M2.1 高速推理版本，适合低延迟场景",
"contextWindow": 204800,
"maxOutputTokens": 131072,
@@ -161,22 +119,16 @@
"tool_use",
"fast"
],
"inputPrice": 4.2,
"inputPrice": 2.1,
"outputPrice": 16.8,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {
"cacheReadPrice": 0.21,
"cacheWritePrice": 2.625,
"outputSpeedTps": 100
}
"defaultTopP": 1,
"extra": {}
},
{
"modelName": "MiniMax-Text-01",
"displayName": "MiniMax Text 01",
"serviceType": [
"chat"
],
"serviceType": ["chat"],
"description": "MiniMax 百万级长上下文文本模型，擅长长文档处理",
"contextWindow": 1000000,
"maxOutputTokens": 131072,
@@ -188,8 +140,8 @@
],
"inputPrice": 1.4,
"outputPrice": 7.7,
"defaultTemperature": 0.1,
"defaultTopP": 0.95,
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {}
}
]

39
compute/providers/mistral.json
View File

@@ -9,17 +9,12 @@
"enabled": false,
"status": "unconfigured",
"priceCurrency": "USD",
"services": [
"chat",
"fast"
],
"services": ["chat", "fast"],
"models": [
{
"modelName": "mistral-large-latest",
"displayName": "Mistral Large 3",
"serviceType": [
"chat"
],
"serviceType": ["chat"],
"description": "Mistral 旗舰模型，256K 上下文，支持视觉和工具调用",
"contextWindow": 256000,
"maxOutputTokens": 262144,
@@ -31,20 +26,18 @@
"tool_use",
"long_context"
],
"inputPrice": 0.5,
"outputPrice": 1.5,
"defaultTemperature": null,
"inputPrice": 0.50,
"outputPrice": 1.50,
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {}
},
{
"modelName": "mistral-small-latest",
"displayName": "Mistral Small 3.2",
"serviceType": [
"fast"
],
"serviceType": ["fast"],
"description": "Mistral 高效小模型，低延迟，适合快速推理场景",
"contextWindow": 128000,
"contextWindow": 130000,
"maxOutputTokens": 8192,
"capabilities": [
"chat",
@@ -54,28 +47,26 @@
"fast",
"tool_use"
],
"inputPrice": 0.1,
"outputPrice": 0.3,
"defaultTemperature": null,
"inputPrice": 0.10,
"outputPrice": 0.30,
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {}
},
{
"modelName": "codestral-latest",
"displayName": "Codestral",
"serviceType": [
"chat"
],
"serviceType": ["chat"],
"description": "Mistral 专业代码模型，支持 Fill-in-the-Middle 补全",
"contextWindow": 128000,
"contextWindow": 256000,
"maxOutputTokens": 32768,
"capabilities": [
"code",
"fill_in_the_middle"
],
"inputPrice": 0.3,
"outputPrice": 0.9,
"defaultTemperature": null,
"inputPrice": 0.30,
"outputPrice": 0.90,
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {}
}

139
compute/providers/moonshot.json
View File

@@ -14,64 +14,13 @@
"reasoning"
],
"models": [
{
"modelName": "kimi-k2.6",
"displayName": "Kimi K2.6",
"serviceType": [
"chat"
],
"description": "月之暗面 Kimi K2.6，最新多模态模型，长程代码编写与 Agent 自主执行能力增强，256K 上下文，支持思考/非思考模式",
"contextWindow": 256000,
"maxOutputTokens": 32768,
"capabilities": [
"chat",
"reasoning",
"code",
"tool_use",
"agent",
"long_context",
"vision",
"video_understanding",
"image_understanding"
],
"inputPrice": 6.5,
"outputPrice": 27,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {
"thinking": {
"default": "enabled",
"disabledTemperature": 0.6
},
"cacheHitPrice": 1.1,
"supportedImageFormats": [
"png",
"jpeg",
"webp",
"gif"
],
"supportedVideoFormats": [
"mp4",
"mpeg",
"mov",
"avi",
"x-flv",
"mpg",
"webm",
"wmv",
"3gpp"
]
}
},
{
"modelName": "kimi-k2.5",
"displayName": "Kimi K2.5",
"serviceType": [
"chat"
],
"description": "月之暗面Kimi K2.5，原生多模态模型，256K 上下文，支持思考/非思考模式",
"serviceType": ["chat"],
"description": "月之暗面Kimi K2.5，2026年1月发布的原生多模态模型，支持视觉编码和智能体集群",
"contextWindow": 256000,
"maxOutputTokens": 32768,
"maxOutputTokens": 16384,
"capabilities": [
"chat",
"reasoning",
@@ -83,40 +32,15 @@
],
"inputPrice": 4,
"outputPrice": 21,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {
"thinking": {
"default": "enabled",
"disabledTemperature": 0.6
},
"cacheHitPrice": 0.7,
"supportedImageFormats": [
"png",
"jpeg",
"webp",
"gif"
],
"supportedVideoFormats": [
"mp4",
"mpeg",
"mov",
"avi",
"x-flv",
"mpg",
"webm",
"wmv",
"3gpp"
]
}
"defaultTemperature": 0.7,
"defaultTopP": 0.9,
"extra": {}
},
{
"modelName": "kimi-k2",
"displayName": "Kimi K2",
"serviceType": [
"chat"
],
"description": "月之暗面Kimi K2，万亿参数MoE模型(32B激活)，256K上下文，专为智能体设计",
"serviceType": ["chat"],
"description": "月之暗面Kimi K2，万亿参数MoE模型(320B激活)，256K上下文，专为智能体设计",
"contextWindow": 256000,
"maxOutputTokens": 8192,
"capabilities": [
@@ -125,22 +49,19 @@
"code",
"tool_use",
"agent",
"long_context"
"long_context",
"vision"
],
"inputPrice": 4,
"outputPrice": 16,
"defaultTemperature": 0.6,
"defaultTopP": 1,
"extra": {
"cacheHitPrice": 1
}
"defaultTemperature": 0.7,
"defaultTopP": 0.9,
"extra": {}
},
{
"modelName": "kimi-k2-thinking",
"displayName": "Kimi K2 思考版",
"serviceType": [
"reasoning"
],
"serviceType": ["reasoning"],
"description": "月之暗面Kimi K2思考版，256K上下文，深度推理能力",
"contextWindow": 256000,
"maxOutputTokens": 16384,
@@ -153,18 +74,14 @@
],
"inputPrice": 4,
"outputPrice": 16,
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {
"cacheHitPrice": 1
}
"defaultTemperature": 0.7,
"defaultTopP": 0.9,
"extra": {}
},
{
"modelName": "moonshot-v1-8k",
"displayName": "Moonshot V1 8K",
"serviceType": [
"chat"
],
"serviceType": ["chat"],
"description": "月之暗面标准模型，8K上下文",
"contextWindow": 8192,
"maxOutputTokens": 4096,
@@ -174,16 +91,14 @@
],
"inputPrice": 2,
"outputPrice": 2,
"defaultTemperature": 0,
"defaultTopP": 1,
"defaultTemperature": 0.7,
"defaultTopP": 0.9,
"extra": {}
},
{
"modelName": "moonshot-v1-32k",
"displayName": "Moonshot V1 32K",
"serviceType": [
"chat"
],
"serviceType": ["chat"],
"description": "月之暗面标准模型，32K上下文",
"contextWindow": 32768,
"maxOutputTokens": 4096,
@@ -194,16 +109,14 @@
],
"inputPrice": 5,
"outputPrice": 5,
"defaultTemperature": 0,
"defaultTopP": 1,
"defaultTemperature": 0.7,
"defaultTopP": 0.9,
"extra": {}
},
{
"modelName": "moonshot-v1-128k",
"displayName": "Moonshot V1 128K",
"serviceType": [
"chat"
],
"serviceType": ["chat"],
"description": "月之暗面标准模型，128K上下文",
"contextWindow": 131072,
"maxOutputTokens": 4096,
@@ -214,8 +127,8 @@
],
"inputPrice": 10,
"outputPrice": 10,
"defaultTemperature": 0,
"defaultTopP": 1,
"defaultTemperature": 0.7,
"defaultTopP": 0.9,
"extra": {}
}
]

3
compute/providers/ollama.json
View File

@@ -19,14 +19,11 @@
"serviceType": ["chat"],
"description": "本地运行的 Llama 3.1 70B",
"contextWindow": 131072,
"maxOutputTokens": 8192,
"capabilities": [
"chat",
"code",
"reasoning"
],
"defaultTemperature": 0.8,
"defaultTopP": 0.9,
"extra": {}
}
]

190
compute/providers/openai.json
View File

@@ -25,9 +25,7 @@
{
"modelName": "gpt-5.2",
"displayName": "GPT-5.2",
"serviceType": [
"chat"
],
"serviceType": ["chat"],
"description": "OpenAI 最新旗舰模型，400K 上下文，编码和智能体任务优化",
"contextWindow": 400000,
"maxOutputTokens": 128000,
@@ -48,9 +46,7 @@
{
"modelName": "gpt-5.2-pro",
"displayName": "GPT-5.2 Pro",
"serviceType": [
"chat"
],
"serviceType": ["chat"],
"description": "OpenAI GPT-5.2 Pro 专业版，最高性能",
"contextWindow": 400000,
"maxOutputTokens": 128000,
@@ -71,9 +67,7 @@
{
"modelName": "gpt-5.1",
"displayName": "GPT-5.1",
"serviceType": [
"chat"
],
"serviceType": ["chat"],
"description": "OpenAI GPT-5.1，400K 上下文，高性能旗舰",
"contextWindow": 400000,
"maxOutputTokens": 128000,
@@ -94,9 +88,7 @@
{
"modelName": "gpt-5",
"displayName": "GPT-5",
"serviceType": [
"chat"
],
"serviceType": ["chat"],
"description": "OpenAI GPT-5 新一代旗舰模型，统一所有模型能力",
"contextWindow": 400000,
"maxOutputTokens": 128000,
@@ -117,12 +109,10 @@
{
"modelName": "gpt-5-pro",
"displayName": "GPT-5 Pro",
"serviceType": [
"chat"
],
"description": "OpenAI GPT-5 Pro 专业版高性能推理模型",
"serviceType": ["chat"],
"description": "OpenAI GPT-5 Pro 专业版高性能模型",
"contextWindow": 400000,
"maxOutputTokens": 272000,
"maxOutputTokens": 128000,
"capabilities": [
"chat",
"reasoning",
@@ -140,12 +130,10 @@
{
"modelName": "gpt-5-mini",
"displayName": "GPT-5-mini",
"serviceType": [
"chat"
],
"description": "高性价比模型，适合日常对话，400K 上下文",
"contextWindow": 400000,
"maxOutputTokens": 128000,
"serviceType": ["chat"],
"description": "高性价比模型，适合日常对话",
"contextWindow": 128000,
"maxOutputTokens": 32768,
"capabilities": [
"chat",
"fast",
@@ -162,19 +150,17 @@
{
"modelName": "gpt-5-nano",
"displayName": "GPT-5-nano",
"serviceType": [
"fast"
],
"description": "极速响应模型，适合简单任务，400K 上下文",
"contextWindow": 400000,
"maxOutputTokens": 128000,
"serviceType": ["fast"],
"description": "极速响应模型，适合简单任务",
"contextWindow": 128000,
"maxOutputTokens": 16384,
"capabilities": [
"chat",
"code",
"fast"
],
"inputPrice": 0.05,
"outputPrice": 0.4,
"inputPrice": 0.02,
"outputPrice": 0.08,
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {}
@@ -182,11 +168,9 @@
{
"modelName": "gpt-4.1",
"displayName": "GPT-4.1",
"serviceType": [
"chat"
],
"serviceType": ["chat"],
"description": "OpenAI GPT-4.1，支持 1M 上下文，编码和指令遵循能力大幅提升",
"contextWindow": 1047576,
"contextWindow": 1000000,
"maxOutputTokens": 32768,
"capabilities": [
"chat",
@@ -196,8 +180,8 @@
"ultra_long_context",
"tool_use"
],
"inputPrice": 2,
"outputPrice": 8,
"inputPrice": 2.00,
"outputPrice": 8.00,
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {}
@@ -205,11 +189,9 @@
{
"modelName": "gpt-4.1-mini",
"displayName": "GPT-4.1 mini",
"serviceType": [
"chat"
],
"serviceType": ["chat"],
"description": "OpenAI GPT-4.1 mini，1M 上下文，高性价比",
"contextWindow": 1047576,
"contextWindow": 1000000,
"maxOutputTokens": 32768,
"capabilities": [
"chat",
@@ -218,8 +200,8 @@
"ultra_long_context",
"tool_use"
],
"inputPrice": 0.4,
"outputPrice": 1.6,
"inputPrice": 0.40,
"outputPrice": 1.60,
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {}
@@ -227,11 +209,9 @@
{
"modelName": "gpt-4.1-nano",
"displayName": "GPT-4.1 nano",
"serviceType": [
"fast"
],
"serviceType": ["fast"],
"description": "OpenAI GPT-4.1 nano，1M 上下文，极致性价比",
"contextWindow": 1047576,
"contextWindow": 1000000,
"maxOutputTokens": 32768,
"capabilities": [
"chat",
@@ -239,8 +219,8 @@
"ultra_long_context",
"fast"
],
"inputPrice": 0.1,
"outputPrice": 0.4,
"inputPrice": 0.10,
"outputPrice": 0.40,
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {}
@@ -248,9 +228,7 @@
{
"modelName": "gpt-4o",
"displayName": "GPT-4o",
"serviceType": [
"vision"
],
"serviceType": ["vision"],
"description": "多模态模型，支持图像理解",
"contextWindow": 128000,
"maxOutputTokens": 16384,
@@ -260,8 +238,8 @@
"image_understanding",
"tool_use"
],
"inputPrice": 2.5,
"outputPrice": 10,
"inputPrice": 2.50,
"outputPrice": 10.00,
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {}
@@ -269,9 +247,7 @@
{
"modelName": "gpt-4o-mini",
"displayName": "GPT-4o mini",
"serviceType": [
"vision"
],
"serviceType": ["vision"],
"description": "OpenAI GPT-4o mini，高性价比多模态模型",
"contextWindow": 128000,
"maxOutputTokens": 16384,
@@ -282,7 +258,7 @@
"tool_use"
],
"inputPrice": 0.15,
"outputPrice": 0.6,
"outputPrice": 0.60,
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {}
@@ -290,50 +266,35 @@
{
"modelName": "text-embedding-3-small",
"displayName": "text-embedding-3-small",
"serviceType": [
"embedding"
],
"serviceType": ["embedding"],
"description": "轻量文本向量化模型，1536 维",
"contextWindow": 8192,
"capabilities": [
"text_embedding",
"semantic_search"
],
"inputPrice": 0.02,
"defaultTemperature": null,
"defaultTopP": null,
"extra": {}
},
{
"modelName": "text-embedding-3-large",
"displayName": "text-embedding-3-large",
"serviceType": [
"embedding"
],
"serviceType": ["embedding"],
"description": "高精度文本向量化模型，3072 维",
"contextWindow": 8192,
"capabilities": [
"text_embedding",
"semantic_search"
],
"inputPrice": 0.13,
"defaultTemperature": null,
"defaultTopP": null,
"extra": {}
},
{
"modelName": "tts-1",
"displayName": "TTS-1",
"serviceType": [
"tts"
],
"serviceType": ["tts"],
"description": "标准语音合成",
"capabilities": [
"tts"
],
"inputPrice": 15,
"defaultTemperature": null,
"defaultTopP": null,
"extra": {
"voices": [
"alloy",
@@ -354,16 +315,11 @@
{
"modelName": "tts-1-hd",
"displayName": "TTS-1-HD",
"serviceType": [
"tts"
],
"serviceType": ["tts"],
"description": "高清语音合成",
"capabilities": [
"tts"
],
"inputPrice": 30,
"defaultTemperature": null,
"defaultTopP": null,
"extra": {
"voices": [
"alloy",
@@ -384,17 +340,12 @@
{
"modelName": "whisper-1",
"displayName": "Whisper",
"serviceType": [
"asr"
],
"serviceType": ["asr"],
"description": "通用语音识别",
"capabilities": [
"asr",
"multilingual"
],
"inputPrice": 0.006,
"defaultTemperature": null,
"defaultTopP": null,
"extra": {
"maxAudioLength": 600,
"supportedInputFormats": [
@@ -412,9 +363,7 @@
{
"modelName": "o3",
"displayName": "o3",
"serviceType": [
"responses"
],
"serviceType": ["responses"],
"description": "OpenAI o3 推理模型，200K 上下文，支持工具调用",
"contextWindow": 200000,
"maxOutputTokens": 100000,
@@ -426,8 +375,8 @@
"science",
"tool_use"
],
"inputPrice": 2,
"outputPrice": 8,
"inputPrice": 2.00,
"outputPrice": 8.00,
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {}
@@ -435,9 +384,7 @@
{
"modelName": "o3-pro",
"displayName": "o3-pro",
"serviceType": [
"responses"
],
"serviceType": ["responses"],
"description": "OpenAI o3-pro 高级推理模型，200K 上下文",
"contextWindow": 200000,
"maxOutputTokens": 100000,
@@ -448,8 +395,8 @@
"math",
"science"
],
"inputPrice": 20,
"outputPrice": 80,
"inputPrice": 20.00,
"outputPrice": 80.00,
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {}
@@ -457,11 +404,9 @@
{
"modelName": "o3-mini",
"displayName": "o3-mini",
"serviceType": [
"responses"
],
"description": "OpenAI o3-mini 推理模型，200K 上下文",
"contextWindow": 200000,
"serviceType": ["responses"],
"description": "OpenAI o3-mini 推理模型，128K 上下文",
"contextWindow": 128000,
"maxOutputTokens": 100000,
"capabilities": [
"reasoning",
@@ -469,8 +414,8 @@
"code",
"fast"
],
"inputPrice": 1.1,
"outputPrice": 4.4,
"inputPrice": 1.10,
"outputPrice": 4.40,
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {}
@@ -478,9 +423,7 @@
{
"modelName": "o4-mini",
"displayName": "o4-mini",
"serviceType": [
"responses"
],
"serviceType": ["responses"],
"description": "OpenAI o4-mini 推理模型，200K 上下文，支持工具调用",
"contextWindow": 200000,
"maxOutputTokens": 100000,
@@ -491,8 +434,8 @@
"tool_use",
"vision"
],
"inputPrice": 1.1,
"outputPrice": 4.4,
"inputPrice": 1.10,
"outputPrice": 4.40,
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {}
@@ -500,16 +443,12 @@
{
"modelName": "dall-e-3",
"displayName": "DALL-E 3",
"serviceType": [
"image_gen"
],
"serviceType": ["image_gen"],
"description": "高质量文生图模型",
"capabilities": [
"image_generation",
"text_rendering"
],
"defaultTemperature": null,
"defaultTopP": null,
"extra": {
"supportedImageSizes": [
"1024x1024",
@@ -525,12 +464,9 @@
{
"modelName": "gpt-4o-realtime",
"displayName": "GPT-4o Realtime",
"serviceType": [
"omni"
],
"serviceType": ["omni"],
"description": "全能实时模型，支持语音 + 文本 + 视觉",
"contextWindow": 32000,
"maxOutputTokens": 4096,
"contextWindow": 128000,
"capabilities": [
"chat",
"vision",
@@ -538,10 +474,6 @@
"asr",
"realtime"
],
"inputPrice": 4,
"outputPrice": 16,
"defaultTemperature": 0.7,
"defaultTopP": null,
"extra": {
"supportedModes": [
"speech-to-speech",
@@ -553,21 +485,13 @@
{
"modelName": "gpt-4o-realtime-preview",
"displayName": "GPT-4o Realtime Preview",
"serviceType": [
"realtime_voice"
],
"serviceType": ["realtime_voice"],
"description": "实时语音交互模型",
"contextWindow": 32000,
"maxOutputTokens": 4096,
"capabilities": [
"realtime",
"speech_to_speech",
"low_latency"
],
"inputPrice": 5,
"outputPrice": 20,
"defaultTemperature": 0.7,
"defaultTopP": null,
"extra": {
"supportedModes": [
"speech-to-speech",

21
compute/providers/openrouter.json
View File

@@ -16,7 +16,7 @@
"displayName": "OpenRouter Auto",
"serviceType": ["chat"],
"description": "OpenRouter 自动路由，智能选择最优模型",
"contextWindow": 2000000,
"contextWindow": 200000,
"maxOutputTokens": 16384,
"capabilities": [
"chat",
@@ -29,16 +29,19 @@
"extra": {}
},
{
"modelName": "openai/gpt-oss-120b:free",
"displayName": "GPT-OSS 120B (免费)",
"modelName": "google/gemini-2.5-flash-exp:free",
"displayName": "Gemini 2.5 Flash (免费)",
"serviceType": ["chat"],
"description": "通过 OpenRouter 免费使用的 OpenAI GPT-OSS 120B",
"contextWindow": 131072,
"maxOutputTokens": 131072,
"description": "通过 OpenRouter 免费使用的 Gemini 2.5 Flash，1M 上下文",
"contextWindow": 1048576,
"maxOutputTokens": 65535,
"capabilities": [
"chat",
"reasoning",
"code"
"vision",
"tool_use",
"fast",
"long_context"
],
"inputPrice": 0.00,
"outputPrice": 0.00,
@@ -51,8 +54,8 @@
"displayName": "Qwen3 Coder 480B (免费)",
"serviceType": ["chat"],
"description": "通过 OpenRouter 免费使用的 Qwen3 Coder 480B",
"contextWindow": 262000,
"maxOutputTokens": 262000,
"contextWindow": 262144,
"maxOutputTokens": 32768,
"capabilities": [
"code",
"reasoning",

39
compute/providers/perplexity.json
View File

@@ -26,16 +26,9 @@
],
"inputPrice": 3.00,
"outputPrice": 15.00,
"defaultTemperature": null,
"defaultTopP": null,
"extra": {
"requestPricingPer1k": {
"low": 6,
"medium": 10,
"high": 14
},
"pricingNotes": "Total Sonar API cost includes token costs plus a request fee based on search context size."
}
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {}
},
{
"modelName": "sonar-reasoning-pro",
@@ -52,16 +45,9 @@
],
"inputPrice": 2.00,
"outputPrice": 8.00,
"defaultTemperature": null,
"defaultTopP": null,
"extra": {
"requestPricingPer1k": {
"low": 6,
"medium": 10,
"high": 14
},
"pricingNotes": "Total Sonar API cost includes token costs plus a request fee based on search context size."
}
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {}
},
{
"modelName": "sonar",
@@ -78,16 +64,9 @@
],
"inputPrice": 1.00,
"outputPrice": 1.00,
"defaultTemperature": null,
"defaultTopP": null,
"extra": {
"requestPricingPer1k": {
"low": 5,
"medium": 8,
"high": 12
},
"pricingNotes": "Total Sonar API cost includes token costs plus a request fee based on search context size."
}
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {}
}
]
}

26
compute/providers/siliconflow.json
View File

@@ -15,41 +15,41 @@
"modelName": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
"displayName": "Qwen3 Coder 480B (硅基)",
"serviceType": ["chat"],
"description": "Qwen3-Coder-480B-A35B-Instruct，480B MoE（35B 激活），原生 256K 上下文，可通过外推扩展至 1M，面向 Agentic Coding",
"description": "开源最强代码模型，480B MoE 架构，擅长代码生成和 Agent 场景",
"contextWindow": 262144,
"maxOutputTokens": 262144,
"maxOutputTokens": 32768,
"capabilities": [
"chat",
"code",
"long_context",
"reasoning",
"agent",
"tool_use"
],
"inputPrice": 8,
"outputPrice": 16,
"defaultTemperature": null,
"defaultTopP": 0.7,
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {}
},
{
"modelName": "Qwen/Qwen3-235B-A22B-Instruct-2507",
"modelName": "Qwen/Qwen3-235B-A22B-Instruct",
"displayName": "Qwen3 235B (硅基)",
"serviceType": ["chat"],
"description": "Qwen3-235B-A22B-Instruct-2507，235B MoE（22B 激活）通用指令模型，262K 上下文",
"description": "Qwen3 旗舰通用模型，235B MoE 架构，多语言能力突出",
"contextWindow": 262144,
"maxOutputTokens": 262144,
"maxOutputTokens": 38912,
"capabilities": [
"chat",
"reasoning",
"code",
"multilingual",
"tool_use",
"agent",
"long_context"
"vision"
],
"inputPrice": 2.5,
"outputPrice": 10,
"defaultTemperature": null,
"defaultTopP": 0.7,
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {}
},
{
@@ -65,8 +65,6 @@
],
"inputPrice": 0,
"outputPrice": 0,
"defaultTemperature": null,
"defaultTopP": null,
"extra": {}
}
]

78
compute/providers/tencent.json
View File

@@ -9,19 +9,14 @@
"enabled": false,
"status": "unconfigured",
"priceCurrency": "CNY",
"services": [
"chat",
"reasoning"
],
"services": ["chat", "reasoning"],
"models": [
{
"modelName": "hunyuan-2.0-thinking-20251109",
"displayName": "混元 2.0 Think",
"serviceType": [
"reasoning"
],
"serviceType": ["reasoning"],
"description": "腾讯混元最新推理模型，128K输入/64K输出，MoE架构406B总参数",
"contextWindow": 196608,
"contextWindow": 131072,
"maxOutputTokens": 65536,
"capabilities": [
"chat",
@@ -31,65 +26,16 @@
],
"inputPrice": 3.975,
"outputPrice": 15.9,
"defaultTemperature": null,
"defaultTopP": null,
"extra": {
"pricingTiers": [
{
"maxInputTokens": 32768,
"inputPrice": 3.975,
"outputPrice": 15.9
},
{
"maxInputTokens": 131072,
"inputPrice": 5.3,
"outputPrice": 21.2
}
]
}
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {}
},
{
"modelName": "hunyuan-2.0-instruct-20251111",
"displayName": "混元 2.0 Instruct",
"serviceType": [
"chat"
],
"description": "腾讯混元 2.0 指令模型，最大输入128K/最大输出16K，支持联网搜索和 Function Calling",
"contextWindow": 147456,
"maxOutputTokens": 16384,
"capabilities": [
"chat",
"code",
"long_context",
"tool_use"
],
"inputPrice": 3.18,
"outputPrice": 7.95,
"defaultTemperature": null,
"defaultTopP": null,
"extra": {
"pricingTiers": [
{
"maxInputTokens": 32768,
"inputPrice": 3.18,
"outputPrice": 7.95
},
{
"maxInputTokens": 131072,
"inputPrice": 4.505,
"outputPrice": 11.13
}
]
}
},
{
"modelName": "hunyuan-turbos-latest",
"modelName": "hunyuan-turbo-s",
"displayName": "混元 Turbo S",
"serviceType": [
"chat"
],
"description": "腾讯混元高速模型，32K输入/16K输出，支持推理和工具调用",
"contextWindow": 32768,
"serviceType": ["chat"],
"description": "腾讯混元高速模型，262K 上下文，支持推理和工具调用",
"contextWindow": 262144,
"maxOutputTokens": 16384,
"capabilities": [
"chat",
@@ -101,8 +47,8 @@
],
"inputPrice": 0.8,
"outputPrice": 2,
"defaultTemperature": null,
"defaultTopP": null,
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {}
}
]

148
compute/providers/volcengine.json
View File

@@ -23,12 +23,10 @@
{
"modelName": "doubao-2.0-pro",
"displayName": "豆包 2.0 Pro",
"serviceType": [
"chat"
],
"serviceType": ["chat"],
"description": "火山引擎豆包2.0旗舰模型，256K上下文，相比Gemini 3 Pro和GPT 5.2有较大成本优势",
"contextWindow": 256000,
"maxOutputTokens": 128000,
"maxOutputTokens": 16384,
"capabilities": [
"chat",
"reasoning",
@@ -39,19 +37,17 @@
],
"inputPrice": 3.2,
"outputPrice": 16,
"defaultTemperature": 1,
"defaultTopP": 0.7,
"defaultTemperature": 0.7,
"defaultTopP": 0.9,
"extra": {}
},
{
"modelName": "doubao-seed-1.8",
"displayName": "豆包 Seed-1.8",
"serviceType": [
"chat"
],
"serviceType": ["chat"],
"description": "火山引擎豆包1.8旗舰模型，256K上下文，支持深度思考和多模态",
"contextWindow": 256000,
"maxOutputTokens": 16000,
"maxOutputTokens": 16384,
"capabilities": [
"chat",
"reasoning",
@@ -62,19 +58,17 @@
],
"inputPrice": 0.8,
"outputPrice": 2,
"defaultTemperature": 1,
"defaultTopP": 0.7,
"defaultTemperature": 0.7,
"defaultTopP": 0.9,
"extra": {}
},
{
"modelName": "doubao-seed-1.6",
"displayName": "豆包 Seed-1.6",
"serviceType": [
"chat"
],
"serviceType": ["chat"],
"description": "火山引擎豆包1.6旗舰模型，256K上下文",
"contextWindow": 256000,
"maxOutputTokens": 32000,
"maxOutputTokens": 16384,
"capabilities": [
"chat",
"reasoning",
@@ -85,19 +79,17 @@
],
"inputPrice": 0.8,
"outputPrice": 8,
"defaultTemperature": 1,
"defaultTopP": 0.7,
"defaultTemperature": 0.7,
"defaultTopP": 0.9,
"extra": {}
},
{
"modelName": "doubao-seed-1.6-thinking",
"displayName": "豆包 Seed-1.6 思考版",
"serviceType": [
"chat"
],
"serviceType": ["chat"],
"description": "火山引擎豆包1.6深度思考模型，256K上下文",
"contextWindow": 256000,
"maxOutputTokens": 16000,
"maxOutputTokens": 32768,
"capabilities": [
"reasoning",
"math",
@@ -107,19 +99,16 @@
],
"inputPrice": 0.8,
"outputPrice": 8,
"defaultTemperature": 1,
"defaultTopP": 0.7,
"defaultTemperature": 0.3,
"extra": {}
},
{
"modelName": "doubao-seed-1.6-flash",
"displayName": "豆包 Seed-1.6 Flash",
"serviceType": [
"chat"
],
"serviceType": ["chat"],
"description": "火山引擎豆包1.6快速版，成本较前代下降63%",
"contextWindow": 256000,
"maxOutputTokens": 16000,
"maxOutputTokens": 8192,
"capabilities": [
"chat",
"code",
@@ -128,19 +117,17 @@
],
"inputPrice": 0.15,
"outputPrice": 1.5,
"defaultTemperature": 1,
"defaultTopP": 0.7,
"defaultTemperature": 0.7,
"defaultTopP": 0.9,
"extra": {}
},
{
"modelName": "doubao-seed-1.6-lite",
"displayName": "豆包 Seed-1.6 Lite",
"serviceType": [
"chat"
],
"serviceType": ["chat"],
"description": "火山引擎豆包1.6轻量版，极致性价比",
"contextWindow": 256000,
"maxOutputTokens": 32000,
"contextWindow": 128000,
"maxOutputTokens": 4096,
"capabilities": [
"chat",
"code",
@@ -148,19 +135,17 @@
],
"inputPrice": 0.3,
"outputPrice": 0.6,
"defaultTemperature": 1,
"defaultTopP": 0.7,
"defaultTemperature": 0.7,
"defaultTopP": 0.9,
"extra": {}
},
{
"modelName": "doubao-seed-1.6-vision",
"displayName": "豆包 Seed-1.6 视觉版",
"serviceType": [
"vision"
],
"serviceType": ["vision"],
"description": "火山引擎豆包1.6视觉模型，256K上下文，支持视频理解",
"contextWindow": 256000,
"maxOutputTokens": 32000,
"maxOutputTokens": 16384,
"capabilities": [
"chat",
"vision",
@@ -170,16 +155,14 @@
],
"inputPrice": 0.8,
"outputPrice": 8,
"defaultTemperature": 1,
"defaultTopP": 0.7,
"defaultTemperature": 0.7,
"defaultTopP": 0.9,
"extra": {}
},
{
"modelName": "doubao-seed-code",
"displayName": "豆包 Seed-Code",
"serviceType": [
"chat"
],
"serviceType": ["chat"],
"description": "火山引擎豆包编程专用模型，256K上下文",
"contextWindow": 256000,
"maxOutputTokens": 32768,
@@ -191,19 +174,16 @@
],
"inputPrice": 1.2,
"outputPrice": 8,
"defaultTemperature": 1,
"defaultTopP": 0.7,
"defaultTemperature": 0.3,
"extra": {}
},
{
"modelName": "deepseek-v3.2",
"displayName": "DeepSeek V3.2 (火山引擎)",
"serviceType": [
"chat"
],
"serviceType": ["chat"],
"description": "火山引擎托管的DeepSeek V3.2，128K上下文",
"contextWindow": 128000,
"maxOutputTokens": 32000,
"maxOutputTokens": 8000,
"capabilities": [
"chat",
"reasoning",
@@ -219,12 +199,10 @@
{
"modelName": "deepseek-r1",
"displayName": "DeepSeek R1 (火山引擎)",
"serviceType": [
"reasoning"
],
"description": "火山引擎托管的DeepSeek R1推理模型，128K上下文",
"contextWindow": 128000,
"maxOutputTokens": 65536,
"serviceType": ["reasoning"],
"description": "火山引擎托管的DeepSeek R1推理模型，64K上下文",
"contextWindow": 64000,
"maxOutputTokens": 32768,
"capabilities": [
"reasoning",
"math",
@@ -234,18 +212,16 @@
],
"inputPrice": 4,
"outputPrice": 16,
"defaultTemperature": null,
"defaultTopP": null,
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {}
},
{
"modelName": "kimi-k2-volcengine",
"displayName": "Kimi K2 (火山引擎)",
"serviceType": [
"chat"
],
"serviceType": ["chat"],
"description": "火山引擎托管的Kimi K2，万亿参数MoE模型，128K上下文",
"contextWindow": 256000,
"contextWindow": 128000,
"maxOutputTokens": 8192,
"capabilities": [
"chat",
@@ -257,19 +233,17 @@
],
"inputPrice": 4,
"outputPrice": 16,
"defaultTemperature": 0.6,
"defaultTopP": 1,
"defaultTemperature": 0.7,
"defaultTopP": 0.9,
"extra": {}
},
{
"modelName": "glm-4-7",
"displayName": "GLM-4.7 (火山引擎)",
"serviceType": [
"chat"
],
"serviceType": ["chat"],
"description": "火山引擎托管的智谱GLM-4.7，200K上下文",
"contextWindow": 200000,
"maxOutputTokens": 128000,
"maxOutputTokens": 200000,
"capabilities": [
"chat",
"reasoning",
@@ -280,16 +254,14 @@
],
"inputPrice": 4,
"outputPrice": 16,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"defaultTemperature": 0.7,
"defaultTopP": 0.9,
"extra": {}
},
{
"modelName": "doubao-embedding",
"displayName": "豆包 Embedding",
"serviceType": [
"embedding"
],
"serviceType": ["embedding"],
"description": "火山引擎豆包标准向量模型，2560维度",
"contextWindow": 4096,
"capabilities": [
@@ -299,24 +271,18 @@
"chinese_optimized"
],
"inputPrice": 0.5,
"defaultTemperature": null,
"defaultTopP": null,
"extra": {}
},
{
"modelName": "volc-mega-tts-clone",
"displayName": "火山声音复刻",
"serviceType": [
"voice_clone"
],
"serviceType": ["voice_clone"],
"description": "火山引擎声音复刻，支持极少样本",
"capabilities": [
"voice_clone",
"chinese_optimized",
"few_shot"
],
"defaultTemperature": null,
"defaultTopP": null,
"extra": {
"maxCloneSamples": 3,
"minCloneDuration": 5
@@ -325,9 +291,7 @@
{
"modelName": "volc-realtime-voice",
"displayName": "火山实时语音",
"serviceType": [
"realtime_voice"
],
"serviceType": ["realtime_voice"],
"description": "火山引擎实时语音交互",
"capabilities": [
"realtime",
@@ -335,8 +299,6 @@
"low_latency",
"chinese_optimized"
],
"defaultTemperature": null,
"defaultTopP": null,
"extra": {
"supportedModes": [
"speech-to-speech"
@@ -347,17 +309,13 @@
{
"modelName": "volc-simultaneous",
"displayName": "火山同声传译",
"serviceType": [
"simultaneous_interpret"
],
"serviceType": ["simultaneous_interpret"],
"description": "火山引擎同声传译，支持中英双向",
"capabilities": [
"simultaneous_interpretation",
"chinese_optimized",
"low_latency"
],
"defaultTemperature": null,
"defaultTopP": null,
"extra": {
"streamingSupported": true
}
@@ -365,17 +323,13 @@
{
"modelName": "volc-translation",
"displayName": "火山翻译",
"serviceType": [
"translation"
],
"serviceType": ["translation"],
"description": "火山引擎机器翻译，支持多语种互译",
"capabilities": [
"translation",
"multilingual",
"chinese_optimized"
],
"defaultTemperature": null,
"defaultTopP": null,
"extra": {}
}
]

30
compute/providers/xai.json
View File

@@ -12,12 +12,12 @@
"services": ["chat", "reasoning"],
"models": [
{
"modelName": "grok-4.20-0309-reasoning",
"displayName": "Grok 4.20 Reasoning",
"modelName": "grok-4-0709",
"displayName": "Grok 4",
"serviceType": ["chat", "reasoning"],
"description": "xAI 旗舰推理模型，支持函数调用与结构化输出",
"contextWindow": 2000000,
"maxOutputTokens": 16384,
"description": "xAI 最新旗舰推理模型，262K 上下文，支持工具调用",
"contextWindow": 262144,
"maxOutputTokens": 8192,
"capabilities": [
"chat",
"reasoning",
@@ -25,17 +25,17 @@
"vision",
"tool_use"
],
"inputPrice": 2.00,
"outputPrice": 6.00,
"defaultTemperature": null,
"defaultTopP": null,
"inputPrice": 3.00,
"outputPrice": 15.00,
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {}
},
{
"modelName": "grok-4-1-fast-reasoning",
"displayName": "Grok 4.1 Fast Reasoning",
"serviceType": ["chat", "reasoning"],
"description": "xAI 高速推理模型，2M 上下文窗口",
"modelName": "grok-4-1-fast",
"displayName": "Grok 4.1 Fast",
"serviceType": ["chat"],
"description": "xAI 超长上下文高速模型，2M token 窗口",
"contextWindow": 2000000,
"maxOutputTokens": 16384,
"capabilities": [
@@ -49,8 +49,8 @@
],
"inputPrice": 0.20,
"outputPrice": 0.50,
"defaultTemperature": null,
"defaultTopP": null,
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {}
}
]

20
compute/providers/xunfei.json
View File

@@ -12,12 +12,12 @@
"services": ["chat", "reasoning"],
"models": [
{
"modelName": "spark-x",
"modelName": "x1",
"displayName": "讯飞星火 X1",
"serviceType": ["reasoning"],
"description": "讯飞深度推理模型（X2），输入64K/输出128K，支持思考模式与函数调用",
"contextWindow": 65536,
"maxOutputTokens": 128000,
"description": "讯飞最新推理模型，基于全国产算力训练，擅长数学和代码推理",
"contextWindow": 32768,
"maxOutputTokens": 16384,
"capabilities": [
"chat",
"reasoning",
@@ -25,10 +25,12 @@
"math",
"code"
],
"defaultTemperature": 1.2,
"defaultTopP": 0.95,
"inputPrice": 11,
"outputPrice": 11,
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {
"pricingNote": "官方公开文档说明模型价格以控制台及实际购买页为准，未公开固定 token 单价"
"pricingNote": "讯飞按token包计费，不区分输入输出"
}
},
{
@@ -44,10 +46,12 @@
"code",
"tool_use"
],
"inputPrice": 60,
"outputPrice": 60,
"defaultTemperature": 1,
"defaultTopP": 1,
"extra": {
"pricingNote": "官方公开文档说明模型价格以控制台及实际购买页为准，未公开固定 token 单价"
"pricingNote": "讯飞按token包计费，不区分输入输出"
}
}
]

15
compute/providers/zhipu-embedding.json
View File

@@ -18,22 +18,15 @@
"modelName": "embedding-3",
"displayName": "智谱 embedding-3",
"serviceType": ["embedding"],
"description": "智谱嵌入模型v3，支持自定义维度；模型上下文窗口 8K，单条输入最多 3072 tokens。走 OpenAI 兼容 /embeddings 端点（Anthropic 协议不提供 embeddings）",
"contextWindow": 8192,
"description": "智谱嵌入模型v3，支持自定义维度，单条最大3072 tokens。走 OpenAI 兼容 /embeddings 端点（Anthropic 协议不提供 embeddings）",
"contextWindow": 3072,
"capabilities": [
"text_embedding",
"semantic_search",
"rag",
"custom_dimensions"
"rag"
],
"inputPrice": 0.5,
"defaultTemperature": null,
"defaultTopP": null,
"extra": {
"maxInputTokensPerItem": 3072,
"maxBatchItems": 64,
"vectorDimensions": "256-2048"
}
"extra": {}
}
]
}

177
compute/providers/zhipu.json
View File

@@ -18,72 +18,10 @@
"embedding-3"
],
"models": [
{
"modelName": "glm-5.1",
"displayName": "GLM-5.1",
"serviceType": [
"chat"
],
"description": "智谱 GLM-5.1 最新旗舰模型，面向长程 Agentic Coding 与 Autonomous Agent 场景，200K 上下文，128K 最大输出",
"contextWindow": 200000,
"maxOutputTokens": 128000,
"capabilities": [
"chat",
"reasoning",
"code",
"multilingual",
"deep_thinking",
"long_context",
"math",
"tool_use",
"agent"
],
"inputPrice": 6,
"outputPrice": 24,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {
"cacheHitPrice": 1.2,
"thinking": {
"default": "enabled"
}
}
},
{
"modelName": "glm-5-turbo",
"displayName": "GLM-5-Turbo",
"serviceType": [
"chat"
],
"description": "智谱 GLM-5-Turbo，面向 OpenClaw 等长链路任务优化，200K 上下文，128K 最大输出",
"contextWindow": 200000,
"maxOutputTokens": 128000,
"capabilities": [
"chat",
"reasoning",
"code",
"deep_thinking",
"long_context",
"tool_use",
"agent"
],
"inputPrice": 5,
"outputPrice": 22,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {
"cacheHitPrice": 1,
"thinking": {
"default": "enabled"
}
}
},
{
"modelName": "glm-5",
"displayName": "GLM-5",
"serviceType": [
"chat"
],
"serviceType": ["chat"],
"description": "智谱GLM-5，744B参数旗舰基座模型，2026年2月发布，面向Agentic Engineering，编程能力接近Claude Opus",
"contextWindow": 200000,
"maxOutputTokens": 128000,
@@ -100,18 +38,14 @@
],
"inputPrice": 4,
"outputPrice": 18,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {
"cacheHitPrice": 0.8
}
"defaultTemperature": 0.95,
"defaultTopP": 0.7,
"extra": {}
},
{
"modelName": "glm-4.7",
"displayName": "GLM-4.7",
"serviceType": [
"chat"
],
"serviceType": ["chat"],
"description": "智谱GLM-4.7，355B参数MoE架构旗舰模型，200K上下文，编程能力对齐Claude Sonnet",
"contextWindow": 200000,
"maxOutputTokens": 128000,
@@ -124,20 +58,16 @@
"long_context",
"tool_use"
],
"inputPrice": 2,
"outputPrice": 8,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {
"cacheHitPrice": 0.5
}
"inputPrice": 4,
"outputPrice": 16,
"defaultTemperature": 0.95,
"defaultTopP": 0.7,
"extra": {}
},
{
"modelName": "glm-4.7-thinking",
"displayName": "GLM-4.7 Thinking",
"serviceType": [
"reasoning"
],
"serviceType": ["reasoning"],
"description": "智谱GLM-4.7深度思考模式，200K上下文，交错式/保留式/轮级思考",
"contextWindow": 200000,
"maxOutputTokens": 128000,
@@ -148,57 +78,19 @@
"deep_thinking",
"long_context"
],
"inputPrice": 2,
"outputPrice": 8,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {
"cacheHitPrice": 0.5,
"aliasOf": "glm-4.7"
}
},
{
"modelName": "glm-5v-turbo",
"displayName": "GLM-5V-Turbo",
"serviceType": [
"vision"
],
"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，200K 上下文，128K 最大输出",
"contextWindow": 200000,
"maxOutputTokens": 128000,
"capabilities": [
"chat",
"vision",
"video_understanding",
"image_understanding",
"file_understanding",
"reasoning",
"code",
"deep_thinking",
"long_context",
"tool_use",
"agent"
],
"inputPrice": 5,
"outputPrice": 22,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {
"cacheHitPrice": 1,
"thinking": {
"default": "enabled"
}
}
"inputPrice": 8,
"outputPrice": 32,
"defaultTemperature": 0.95,
"defaultTopP": 0.7,
"extra": {}
},
{
"modelName": "glm-4.6v",
"displayName": "GLM-4.6V",
"serviceType": [
"vision"
],
"serviceType": ["vision"],
"description": "智谱GLM-4.6V多模态版，106B/12B MoE，支持图像视频理解与工具调用，128K上下文",
"contextWindow": 128000,
"maxOutputTokens": 32768,
"maxOutputTokens": 8192,
"capabilities": [
"chat",
"vision",
@@ -207,37 +99,28 @@
"long_context",
"tool_use"
],
"inputPrice": 1,
"outputPrice": 4,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {
"cacheHitPrice": 0.2
}
"inputPrice": 8,
"outputPrice": 32,
"extra": {}
},
{
"modelName": "glm-4.6",
"displayName": "GLM-4.6",
"serviceType": [
"chat"
],
"description": "智谱GLM-4.6，增强推理能力，200K上下文",
"contextWindow": 200000,
"maxOutputTokens": 128000,
"serviceType": ["chat"],
"description": "智谱GLM-4.6，增强推理能力，128K上下文",
"contextWindow": 128000,
"maxOutputTokens": 8192,
"capabilities": [
"chat",
"reasoning",
"code",
"multilingual",
"deep_thinking"
"deep_thinking",
"vision"
],
"inputPrice": 2,
"outputPrice": 8,
"defaultTemperature": 1,
"defaultTopP": 0.95,
"extra": {
"cacheHitPrice": 0.5
}
"inputPrice": 5,
"outputPrice": 5,
"extra": {}
}
]
}