智力指数榜 · Artificial Analysis
AA Intelligence / Coding / Agentic Index。客观 bench ≠ 用量热度；交叉验证见天梯。 对标 openrouter.ai/rankings。能力验真 → 模型天梯 · 模型目录 · Providers · 实验室 · 官方 API。
数据状态
同步自 OpenRouter · 336 个模型 · source=openrouter · 2026-08-01T17:19:24.146Z
智力指数榜 · Artificial Analysis
AA Intelligence / Coding / Agentic Index。客观 bench ≠ 用量热度；交叉验证见天梯。 打开天梯做多维对比
80 模天梯
| # | 模型 | 智力 | 代码 | Agent | 输入$/M | 输出$/M | 上下文 |
|---|---|---|---|---|---|---|---|
| 🥇 | Claude Opus 5 anthropic/claude-opus-5 | 60.7 | 78.0 | 55.3 | $5.00 | $25.00 | 1.0M |
| 🥈 | Claude Fable 5 anthropic/claude-fable-5 | 59.9 | 76.5 | 52.8 | $10.00 | $50.00 | 1.0M |
| 🥉 | GPT-5.6 Sol openai/gpt-5.6-sol | 58.9 | 77.4 | 54.0 | $5.00 | $30.00 | 1.1M |
| 4 | Kimi K3 moonshotai/kimi-k3 | 57.1 | 76.2 | 50.1 | $3.00 | $15.00 | 1.0M |
| 5 | Claude Opus 4.8 anthropic/claude-opus-4.8 | 55.7 | 74.3 | 47.2 | $5.00 | $25.00 | 1.0M |
| 6 | GPT-5.6 Terra openai/gpt-5.6-terra | 55.0 | 76.7 | 47.4 | $1.00 | $6.00 | 1.1M |
| 7 | GPT-5.5 openai/gpt-5.5 | 54.8 | 74.9 | 44.9 | $5.00 | $30.00 | 1.1M |
| 8 | Grok 4.5 x-ai/grok-4.5 | 53.8 | 72.4 | 45.7 | $2.00 | $6.00 | 500k |
| 9 | Claude Opus 4.7 anthropic/claude-opus-4.7 | 53.5 | 73.6 | 44.4 | $5.00 | $25.00 | 1.0M |
| 10 | Claude Sonnet 5 anthropic/claude-sonnet-5 | 53.4 | 71.5 | 46.7 | $2.00 | $10.00 | 1.0M |
| 11 | GPT-5.4 openai/gpt-5.4 | 51.4 | 71.1 | 41.1 | $2.50 | $15.00 | 1.1M |
| 12 | GPT-5.6 Luna openai/gpt-5.6-luna | 51.2 | 71.4 | 45.6 | $0.10 | $0.60 | 1.1M |
| 13 | GLM 5.2 z-ai/glm-5.2 | 51.1 | 68.8 | 43.1 | $0.72 | $2.25 | 1.0M |
| 14 | Muse Spark 1.1 meta/muse-spark-1.1 | 50.6 | 71.3 | 37.5 | $1.25 | $4.25 | 1.0M |
| 15 | Gemini 3.5 Flash google/gemini-3.5-flash | 50.2 | 70.1 | 37.4 | $1.50 | $9.00 | 1.0M |
| 16 | Gemini 3.6 Flash google/gemini-3.6-flash | 50.1 | 69.2 | 38.7 | $1.50 | $7.50 | 1.0M |
| 17 | DeepSeek V4 Flash 0731 deepseek/deepseek-v4-flash-0731 | 49.9 | 69.1 | 45.7 | $0.14 | $0.28 | 1.0M |
| 18 | Claude Sonnet 4.6 anthropic/claude-sonnet-4.6 | 47.2 | 63.0 | 40.8 | $3.00 | $15.00 | 1.0M |
| 19 | Gemini 3.1 Pro Preview google/gemini-3.1-pro-preview | 46.5 | 68.8 | 21.4 | $2.00 | $12.00 | 1.0M |
| 20 | Qwen3.7 Max qwen/qwen3.7-max | 46.0 | 66.0 | 30.6 | $1.48 | $4.42 | 1.0M |
| 21 | MiniMax M3 minimax/minimax-m3 | 44.4 | 58.6 | 35.4 | $0.30 | $1.20 | 1.0M |
| 22 | DeepSeek V4 Pro deepseek/deepseek-v4-pro | 44.3 | 59.4 | 36.4 | $0.43 | $0.87 | 1.0M |
| 23 | Kimi K2.6 moonshotai/kimi-k2.6 | 44.2 | 61.8 | 30.3 | $0.60 | $3.41 | 262k |
| 24 | MiMo-V2.5-Pro xiaomi/mimo-v2.5-pro | 42.2 | 60.2 | 29.1 | $0.43 | $0.87 | 1.1M |
| 25 | Kimi K2.7 Code moonshotai/kimi-k2.7-code | 41.9 | 60.8 | 29.6 | $0.73 | $3.50 | 262k |
| 26 | Hy3 preview tencent/hy3-preview | 41.2 | 58.8 | 30.7 | $0.06 | $0.21 | 262k |
| 27 | Nex-N2-Pro nex-agi/nex-n2-pro | 41.0 | 59.1 | 31.0 | $0.25 | $1.00 | 262k |
| 28 | Inkling thinkingmachines/inkling | 40.7 | 52.1 | 32.3 | $1.00 | $4.05 | 1.0M |
| 29 | Inkling Small thinkingmachines/inkling-small | 40.2 | 52.9 | 30.8 | $0.50 | $1.20 | 524k |
| 30 | GLM 5.1 z-ai/glm-5.1 | 40.2 | 55.8 | 29.9 | $0.97 | $3.04 | 205k |
| 31 | GPT-5.4 Mini openai/gpt-5.4-mini | 40.0 | 56.1 | 30.2 | $0.75 | $4.50 | 400k |
| 32 | Grok Build 0.1 x-ai/grok-build-0.1 | 39.8 | 51.5 | 28.0 | $1.00 | $2.00 | 256k |
| 33 | Qwen3.6 Plus qwen/qwen3.6-plus | 39.6 | 54.5 | 27.6 | $0.33 | $1.95 | 1.0M |
| 34 | Qwen3.7 Plus qwen/qwen3.7-plus | 39.0 | 55.9 | 20.8 | $0.32 | $1.28 | 1.0M |
| 35 | GPT-5.4 Nano openai/gpt-5.4-nano | 38.2 | 56.1 | 27.5 | $0.20 | $1.25 | 400k |
| 36 | MiniMax M2.7 minimax/minimax-m2.7 | 38.1 | 52.6 | 25.6 | $0.25 | $1.00 | 205k |
| 37 | Nemotron 3 Ultra nvidia/nemotron-3-ultra-550b-a55b | 37.8 | 49.3 | 27.4 | $0.60 | $3.60 | 512k |
| 38 | Nemotron 3 Ultra (free) nvidia/nemotron-3-ultra-550b-a55b:free | 37.8 | 49.3 | 27.4 | $0.00 | $0.00 | 1.0M |
| 39 | Grok 4.3 x-ai/grok-4.3 | 37.6 | 42.2 | 24.1 | $1.25 | $2.50 | 1.0M |
| 40 | MiMo-V2.5 xiaomi/mimo-v2.5 | 37.2 | 56.8 | 23.7 | $0.14 | $0.28 | 1.1M |
| 41 | Qwen3.6 27B qwen/qwen3.6-27b | 37.1 | 53.7 | 27.0 | $0.30 | $2.00 | 262k |
| 42 | GPT-5.1 openai/gpt-5.1 | 36.9 | 49.4 | 21.0 | $1.25 | $10.00 | 400k |
| 43 | Gemini 3.5 Flash Lite google/gemini-3.5-flash-lite | 36.5 | 49.3 | 26.8 | $0.30 | $2.50 | 1.0M |
| 44 | Claude Sonnet 4.5 anthropic/claude-sonnet-4.5 | 36.4 | 52.1 | 24.6 | $3.00 | $15.00 | 1.0M |
| 45 | Kimi K2.5 moonshotai/kimi-k2.5 | 35.4 | 46.8 | 21.7 | $0.57 | $2.85 | 262k |
| 46 | GPT-5 openai/gpt-5 | 34.7 | 37.8 | 25.7 | $1.25 | $10.00 | 400k |
| 47 | KAT-Coder-Pro V2 kwaipilot/kat-coder-pro-v2 | 33.7 | 59.5 | 15.5 | $0.30 | $1.20 | 262k |
| 48 | Qwen3.5 397B A17B qwen/qwen3.5-397b-a17b | 33.7 | 48.2 | 19.8 | $0.39 | $2.34 | 262k |
| 49 | GLM 4.7 z-ai/glm-4.7 | 33.7 | 45.3 | 25.4 | $0.40 | $1.75 | 205k |
| 50 | Qwen3.5-122B-A10B qwen/qwen3.5-122b-a10b | 32.3 | 45.7 | 20.7 | $0.26 | $2.08 | 262k |
| 51 | DeepSeek V3.2 deepseek/deepseek-v3.2 | 32.0 | 44.2 | 18.3 | $0.27 | $0.40 | 164k |
| 52 | Qwen3.6 35B A3B qwen/qwen3.6-35b-a3b | 31.6 | 41.9 | 21.4 | $0.14 | $1.00 | 262k |
| 53 | Ring-2.6-1T inclusionai/ring-2.6-1t | 30.6 | 42.8 | 18.9 | $0.07 | $0.63 | 262k |
| 54 | DeepSeek V3.1 Terminus deepseek/deepseek-v3.1-terminus | 30.4 | 43.5 | 18.1 | $0.27 | $1.00 | 164k |
| 55 | Step 3.7 Flash stepfun/step-3.7-flash | 30.3 | 39.6 | 21.5 | $0.20 | $1.15 | 262k |
| 56 | Mistral Medium 3.5 mistralai/mistral-medium-3-5 | 29.9 | 46.9 | 19.0 | $1.50 | $7.50 | 262k |
| 57 | Claude Haiku 4.5 anthropic/claude-haiku-4.5 | 29.6 | 43.9 | 16.4 | $1.00 | $5.00 | 200k |
| 58 | Gemma 4 31B google/gemma-4-31b-it | 29.4 | 43.4 | 14.4 | $0.10 | $0.34 | 262k |
| 59 | Gemma 4 31B (free) google/gemma-4-31b-it:free | 29.4 | 43.4 | 14.4 | $0.00 | $0.00 | 262k |
| 60 | Claude Sonnet 4 anthropic/claude-sonnet-4 | 28.9 | 37.6 | 16.6 | $3.00 | $15.00 | 1.0M |
| 61 | GLM 4.6 z-ai/glm-4.6 | 28.7 | 45.8 | 17.7 | $0.50 | $2.00 | 205k |
| 62 | Gemini 2.5 Pro google/gemini-2.5-pro | 25.8 | 33.3 | 7.1 | $1.25 | $10.00 | 1.0M |
| 63 | Gemma 4 26B A4B google/gemma-4-26b-a4b-it | 25.7 | 39.3 | 11.0 | $0.07 | $0.34 | 262k |
| 64 | Gemma 4 26B A4B (free) google/gemma-4-26b-a4b-it:free | 25.7 | 39.3 | 11.0 | $0.00 | $0.00 | 262k |
| 65 | Nemotron 3 Super nvidia/nemotron-3-super-120b-a12b | 25.4 | 37.7 | 8.7 | $0.08 | $0.40 | 1.0M |
| 66 | Nemotron 3 Super (free) nvidia/nemotron-3-super-120b-a12b:free | 25.4 | 37.7 | 8.7 | $0.00 | $0.00 | 262k |
| 67 | GPT-5 Mini openai/gpt-5-mini | 25.3 | 15.6 | 19.4 | $0.25 | $2.00 | 400k |
| 68 | Gemini 3.1 Flash Lite Preview google/gemini-3.1-flash-lite-preview | 25.0 | 34.7 | 6.2 | $0.25 | $1.50 | 1.0M |
| 69 | Qwen3.5-35B-A3B qwen/qwen3.5-35b-a3b | 24.0 | 37.0 | 11.8 | $0.14 | $1.00 | 262k |
| 70 | gpt-oss-120b openai/gpt-oss-120b | 23.8 | 30.4 | 13.2 | $0.04 | $0.17 | 131k |
| 71 | Command A cohere/command-a | 22.5 | 27.8 | 9.2 | $2.50 | $10.00 | 256k |
| 72 | Qwen3.5-9B qwen/qwen3.5-9b | 21.4 | 28.7 | 7.4 | $0.10 | $0.15 | 262k |
| 73 | Mercury 2 inception/mercury-2 | 21.4 | 31.1 | 9.6 | $0.25 | $0.75 | 128k |
| 74 | Qwen3 Coder Next qwen/qwen3-coder-next | 21.1 | 36.2 | 8.8 | $0.12 | $0.80 | 262k |
| 75 | North Mini Code (free) cohere/north-mini-code:free | 19.8 | 36.5 | 3.1 | $0.00 | $0.00 | 256k |
| 76 | Mistral Small 4 mistralai/mistral-small-2603 | 19.6 | 26.6 | 4.7 | $0.15 | $0.60 | 262k |
| 77 | Qwen3 235B A22B Thinking 2507 qwen/qwen3-235b-a22b-thinking-2507 | 19.6 | 22.1 | 3.8 | $0.23 | $2.30 | 262k |
| 78 | R1 deepseek/deepseek-r1 | 18.5 | 24.6 | 3.1 | $0.70 | $2.50 | 164k |
| 79 | Trinity Large Thinking arcee-ai/trinity-large-thinking | 18.2 | 25.8 | 3.7 | $0.22 | $0.85 | 262k |
| 80 | Nova 2 Lite amazon/nova-2-lite-v1 | 18.2 | 23.0 | 3.1 | $0.30 | $2.50 | 1.0M |
数据来源 OpenRouter 公开 Rankings API（约小时级缓存）。用量/份额反映路由平台流量，不等于全球真实份额； 智力指数来自 Artificial Analysis 透出字段。选型请结合 天梯、目录与自有压测。
站内推荐
同系工具：Clavue CLI / imux IDE / clavue-2.1
Clavue · CLI 与 Agent 平台
CLI / GUI Agent 运行时、设备登录、会员额度与 OpenAI 兼容 API（api.clavue.com）。开发脚本、CI 与本地工具一条链路。
打开 Clavue ↗imux · 原生 macOS AI IDE
Clavue 平台的原生 IDE：多 Agent 分屏、Ghostty 级终端、Agent Chat、浏览器自动化与 Supervisor——不是又一个 Electron 壳。
了解 imux ↗Clavue 2.1 · 产品级大模型
旗舰模型 clavue-2.1（及 fast / pro / rev）：适合复杂推理与长程 Agent 循环。Chat、imux、API 共用会员额度。
查看模型 ↗