Coding Index
Artificial Analysis's Coding Index: a composite of the publisher's own coding evaluation runs, on a 0 to 100 scale, as OpenRouter carries it. 118 models ranked, as carried in OpenRouter's catalog.
retrieved sha256 · 0bb53a77713fartificialanalysis.ai ↗
The top 26 of 118, 80.7 down to 69.2 on the Coding Index. 7 of them are on the frontier below, in the accent.
Every column stands on zero on the index's own 0 to 100 scale, so a column half as tall is half the score, and the maker's mark hangs under each; the leaderboard below carries every ranked model. Scores are Artificial Analysiss, as OpenRouter carries them.
10 models sit on the frontier: nothing scores higher for less.
- the attractive quadrant
- the frontier
- a priced model, in its maker's tone
- on the frontier, named
- 01Ling 3.0 Tiny26.5 · $0.023 blended
- 02Ling 3.0 Flash50.6 · $0.032 blended · +24.1 at 1.4x
- 03DeepSeek V4 Flash 073169.1 · $0.09 blended · +18.5 at 2.9x
- 04GLM 5.3 Flash71.5 · $0.238 blended · +2.4 at 2.6x
- 05GLM 5.374.8 · $0.402 blended · +3.3 at 1.7x
- 06Gemini 3.7 Flash76.1 · $1.50 blended · +1.3 at 3.7x
- 07Gemini 3.8 Flash76.3 · $1.65 blended · +0.2 at 1.1x
- 08GPT-5.6 Terra76.7 · $4.50 blended · +0.4 at 2.7x
- 09GPT-5.6 Sol78.3 · $8.00 blended · +1.6 at 1.8x
- 10Claude Fable 5.180.7 · $20.00 blended · +2.4 at 2.5x
116 of 118 ranked models carry a list price; 2 are free or unlisted and not in the field. Price is the blended OpenRouter rate, three input tokens to one output, on a logarithmic axis. Each dot is filled in its maker's tone, as the columns above are. The numbered and named dots are the frontier, cheapest first; each step after the first names what it adds over the step before, and at what multiple of the blended price. A hover over any dot names it and its two figures, a click opens the model, and a model named in the picker above keeps its label on. 19 of 116 priced models sit in the attractive quadrant: at or above the median score of 49.3 and at or below the median blended price of $0.734. The tinted corner is that quadrant.
All 118, best first.
Artificial Analysis · 0 to 10012 more authors are on this board, the largest with 3 models. Search a name to reach one.
| Position | Model | Author | Score | In / out per M | Context | Listed |
|---|---|---|---|---|---|---|
| 01 | Claude Fable 5.1 | Anthropic | 80.7 | $10.00 / $50.00 | 1M | |
| 02 | GPT-5.6 Sol | OpenAI | 78.3 | $4.00 / $20.00 | 1.05M | |
| 03 | Claude Opus 5 | Anthropic | 78.0 | $5.00 / $25.00 | 1M | |
| 04 | GPT-6 Astra | OpenAI | 76.9 | $10.00 / $50.00 | 1.05M | |
| 05 | GPT-5.6 Terra | OpenAI | 76.7 | $2.00 / $12.00 | 1.05M | |
| 06 | Claude Fable 5 | Anthropic | 76.5 | $10.00 / $50.00 | 1M | |
| 07 | Gemini 3.8 Flash | 76.3 | $0.825 / $4.13 | 1.05M | ||
| 08 | Qwen3.8 Max (0902) | Qwen | 76.2 | $2.00 / $6.00 | 1M | |
| 09 | Gemini 3.7 Flash | 76.1 | $0.75 / $3.75 | 1.05M | ||
| 10 | Grok 4.6 | SpaceXAI | 75.9 | $2.00 / $6.00 | 500k | |
| 11 | Muse Spark 1.3 | Meta | 75.8 | $1.25 / $4.25 | 1.05M | |
| 12 | GPT-5.5 | OpenAI | 74.9 | $5.00 / $30.00 | 1.05M | |
| 13 | GLM 5.3 | Z.ai | 74.8 | $0.0699 / $1.40 | 1.05M | |
| 14 | Claude Opus 4.8 | Anthropic | 74.3 | $5.00 / $25.00 | 1M | |
| 15 | Claude Opus 4.7 | Anthropic | 73.6 | $5.00 / $25.00 | 1M | |
| 16 | Grok 4.5 | SpaceXAI | 72.4 | $2.00 / $6.00 | 500k | |
| 17 | Muse Spark 1.2 | Meta | 72.2 | $1.25 / $4.25 | 1.05M | |
| 18 | Kimi K3 | MoonshotAI | 72.0 | $0.192 / $16.00 | 1.05M | |
| 19 | Qwen3.8 2.4T A95B | Qwen | 71.9 | $2.00 / $6.00 | 1.05M | |
| 20 | Claude Sonnet 5 | Anthropic | 71.5 | $2.00 / $10.00 | 1M | |
| 21 | GLM 5.3 Flash | Z.ai | 71.5 | $0.15 / $0.50 | 1.05M | |
| 22 | GPT-5.6 Luna | OpenAI | 71.4 | $0.20 / $1.20 | 1.05M | |
| 23 | Muse Spark 1.1 | Meta | 71.3 | $1.25 / $4.25 | 1.05M | |
| 24 | GPT-5.4 | OpenAI | 71.1 | $2.50 / $15.00 | 1.05M | |
| 25 | Gemini 3.5 Flash | 70.1 | $1.50 / $9.00 | 1.05M | ||
| 26 | Gemini 3.6 Flash | 69.2 | $0.75 / $3.75 | 1.05M | ||
| 27 | DeepSeek V4 Flash 0731 | DeepSeek | 69.1 | $0.06 / $0.18 | 1.05M | |
| 28 | DeepSeek V4 Pro 0813 | DeepSeek | 68.8 | $0.111 / $4.44 | 1.05M | |
| 29 | Gemini 3.1 Pro Preview | 68.8 | $2.00 / $12.00 | 1.05M | ||
| 30 | GLM 5.2 | Z.ai | 68.8 | $0.007 / $6.80 | 1.05M | |
| 31 | Qwen3.8 27B | Qwen | 68.1 | $0.15 / $1.88 | 1.02M | |
| 32 | Qwen3.7 Max | Qwen | 66.0 | $1.48 / $4.42 | 1M | |
| 33 | DeepSeek V4 Flash Vision Exp | DeepSeek | 65.0 | $0.44 / $1.32 | 1.05M | |
| 34 | Motif 3 | Motif Technologies | 63.5 | $0.30 / $1.20 | 262k | |
| 35 | Claude Sonnet 4.6 | Anthropic | 63.0 | $3.00 / $15.00 | 1M | |
| 36 | Kimi K2.6 | MoonshotAI | 61.8 | $0.65 / $3.41 | 262k | |
| 37 | Kimi K2.7 Code | MoonshotAI | 60.8 | $0.70 / $3.75 | 262k | |
| 38 | MiMo-V2.5-Pro | Xiaomi | 60.2 | $0.435 / $0.87 | 1.02M | |
| 39 | Hy3 preview | Tencent | 58.8 | $0.18 / $0.60 | 262k | |
| 40 | MiniMax M3 | MiniMax | 58.6 | $0.30 / $1.20 | 1.05M | |
| 41 | Ling 3.0 Flash VL | inclusionAI | 57.0 | $0.06 / $0.18 | 131k | |
| 42 | MiMo-V2.5 | Xiaomi | 56.8 | $0.40 / $2.00 | 1M | |
| 43 | DeepSeek V4 Flash 0423 | DeepSeek | 56.2 | $0.09 / $0.18 | 1.05M | |
| 44 | GPT-5.4 Mini | OpenAI | 56.1 | $0.75 / $4.50 | 400k | |
| 45 | GPT-5.4 Nano | OpenAI | 56.1 | $0.20 / $1.25 | 400k | |
| 46 | Qwen3.7 Plus | Qwen | 55.9 | $0.32 / $1.28 | 1M | |
| 47 | GLM 5.1 | Z.ai | 55.8 | $1.40 / $4.40 | 205k | |
| 48 | Qwen3.6 Plus | Qwen | 54.5 | $0.325 / $1.95 | 1M | |
| 49 | Qwen3.6 27B | Qwen | 53.7 | $0.32 / $3.20 | 262k | |
| 50 | Inkling Small | Thinking Machines | 52.9 | $0.45 / $1.20 | 524k | |
| 51 | Solar Pro 4 | Upstage | 52.7 | $0.15 / $0.60 | 524k | |
| 52 | MiniMax M2.7 | MiniMax | 52.6 | $0.21 / $0.84 | 197k | |
| 53 | Claude Sonnet 4.5 | Anthropic | 52.1 | $3.00 / $15.00 | 1M | |
| 54 | Inkling | Thinking Machines | 52.1 | $1.00 / $4.05 | 524k | |
| 55 | Grok Build 0.1 | SpaceXAI | 51.5 | $1.00 / $2.00 | 256k | |
| 56 | Ling 3.0 Flash | inclusionAI | 50.6 | $0.021 / $0.063 | 262k | |
| 57 | GPT-5.1 | OpenAI | 49.4 | $1.25 / $10.00 | 400k | |
| 58 | Gemini 3.5 Flash Lite | 49.3 | $0.30 / $2.50 | 1.05M | ||
| 59 | Nemotron 3 Ultra | NVIDIA | 49.3 | $0.50 / $2.20 | 262k | |
| 60 | Muse Glimmer 30B | Meta | 49.0 | $0.35 / $1.50 | 131k | |
| 61 | Qwen3.5 397B A17B | Qwen | 48.2 | $0.45 / $3.00 | 262k | |
| 62 | Mistral Medium 3.5 | Mistral | 46.9 | $1.50 / $7.50 | 262k | |
| 63 | Kimi K2.5 | MoonshotAI | 46.8 | $0.50 / $2.50 | 262k | |
| 64 | GLM 4.6 | Z.ai | 45.8 | $0.50 / $2.00 | 205k | |
| 65 | Qwen3.5-122B-A10B | Qwen | 45.7 | $0.26 / $2.08 | 262k | |
| 66 | LongCat 2.0 | Meituan | 45.3 | $0.30 / $1.20 | 1.05M | |
| 67 | GLM 4.7 | Z.ai | 45.3 | $0.40 / $1.75 | 205k | |
| 68 | DeepSeek V3.2 | DeepSeek | 44.2 | $0.26 / $0.38 | 164k | |
| 69 | Claude Haiku 4.5 | Anthropic | 43.9 | $1.00 / $5.00 | 200k | |
| 70 | DeepSeek V3.1 Terminus | DeepSeek | 43.5 | $0.27 / $1.00 | 164k | |
| 71 | Gemma 4 31B | 43.4 | $0.20 / $0.40 | 262k | ||
| 72 | Grok 4.3 | SpaceXAI | 42.2 | $1.25 / $2.50 | 1M | |
| 73 | Qwen3.6 35B A3B | Qwen | 41.9 | $0.10 / $0.90 | 262k | |
| 74 | Step 3.7 Flash | StepFun | 39.6 | $0.20 / $1.15 | 262k | |
| 75 | Gemma 4 26B A4B | 39.3 | $0.09 / $0.30 | 262k | ||
| 76 | GPT-5 | OpenAI | 37.8 | $1.25 / $10.00 | 400k | |
| 77 | Nemotron 3 Super | NVIDIA | 37.7 | $0.075 / $0.40 | 262k | |
| 78 | Qwen3.5-35B-A3B | Qwen | 37.0 | $0.14 / $1.00 | 262k | |
| 79 | North Mini Code (free) | Cohere | 36.5 | free | 256k | |
| 80 | Qwen3 Coder Next | Qwen | 36.2 | $0.12 / $0.80 | 262k | |
| 81 | Gemini 3.1 Flash Lite Preview | 34.7 | $0.25 / $1.50 | 1.05M | ||
| 82 | Gemini 2.5 Pro | 33.3 | $1.25 / $10.00 | 1.05M | ||
| 83 | Devstral 2 2512 | Mistral | 31.3 | $0.40 / $2.00 | 262k | |
| 84 | gpt-oss-120b | OpenAI | 30.4 | $0.15 / $0.60 | 131k | |
| 85 | Qwen3.5-9B | Qwen | 28.7 | $0.17 / $0.25 | 262k | |
| 86 | Command A+ | Cohere | 27.8 | $0.30 / $1.50 | 192k | |
| 87 | Nemotron 3.5 Lightning | NVIDIA | 26.8 | $0.07 / $0.20 | 1.05M | |
| 88 | Mistral Small 4 | Mistral | 26.6 | $0.15 / $0.60 | 262k | |
| 89 | Ling 3.0 Tiny | inclusionAI | 26.5 | $0.015 / $0.045 | 262k | |
| 90 | GPT-4o (2024-05-13) | OpenAI | 24.2 | $5.00 / $15.00 | 128k | |
| 91 | Nova 2 Lite | Amazon | 23.0 | $0.30 / $2.50 | 1M | |
| 92 | Granite 4.2 8B | IBM Granite | 22.4 | $0.06 / $0.15 | 131k | |
| 93 | Qwen3 235B A22B Thinking 2507 | Qwen | 22.1 | $0.45 / $3.50 | 128k | |
| 94 | GPT-4 Turbo (batch) | OpenAI | 21.5 | $5.00 / $15.00 | 128k | |
| 95 | DeepSeek V3 0324 | DeepSeek | 21.2 | $0.29 / $1.14 | 164k | |
| 96 | Kimi K2 Thinking | MoonshotAI | 21.0 | $0.60 / $2.50 | 262k | |
| 97 | gpt-oss-20b | OpenAI | 20.7 | $0.018 / $0.09 | 131k | |
| 98 | Mistral Medium 3.1 | Mistral | 20.5 | $0.44 / $2.20 | 131k | |
| 99 | GPT-4.1 Mini | OpenAI | 20.2 | $0.40 / $1.60 | 1.05M | |
| 100 | Mistral Large 3 2512 | Mistral | 20.1 | $0.50 / $1.50 | 262k | |
| 101 | Llama 4 Maverick | Meta Llama | 16.3 | $0.35 / $1.00 | 524k | |
| 102 | GPT-5 Mini | OpenAI | 15.6 | $0.25 / $2.00 | 400k | |
| 103 | Qwen3 32B | Qwen | 15.3 | $0.08 / $0.28 | 131k | |
| 104 | Ministral 3 14B 2512 | Mistral | 14.4 | $0.20 / $0.20 | 262k | |
| 105 | Nemotron 3 Nano 30B A3B | NVIDIA | 14.4 | $0.06 / $0.24 | 262k | |
| 106 | Nemotron 3 Nano Omni (free) | NVIDIA | 13.8 | free | 256k | |
| 107 | Qwen3 14B | Qwen | 13.8 | $0.12 / $0.24 | 41k | |
| 108 | GPT-4 | OpenAI | 13.1 | $30.00 / $60.00 | 8k | |
| 109 | Llama 3.3 70B Instruct | Meta Llama | 11.9 | $0.71 / $0.71 | 131k | |
| 110 | GPT-4o-mini | OpenAI | 11.4 | $0.15 / $0.60 | 128k | |
| 111 | GPT-4.1 Nano | OpenAI | 11.1 | $0.10 / $0.40 | 1.05M | |
| 112 | Gemma 3 27B | 10.1 | $0.08 / $0.35 | 131k | ||
| 113 | Ministral 3 8B 2512 | Mistral | 9.7 | $0.15 / $0.15 | 262k | |
| 114 | Llama 4 Scout | Meta Llama | 8.2 | $0.10 / $0.30 | 1.31M | |
| 115 | Gemma 3 12B | 5.8 | $0.05 / $0.15 | 131k | ||
| 116 | Llama 3.1 8B Instruct | Meta Llama | 5.4 | $0.02 / $0.04 | 131k | |
| 117 | Ministral 3 3B 2512 | Mistral | 4.8 | $0.10 / $0.10 | 131k | |
| 118 | Gemma 3 4B | 2.7 | $0.05 / $0.10 | 131k |
A composite the publisher computes from its own evaluation runs. It orders models on the publisher's tasks, not yours; a gap of a point or two is within what a different task mix would move. It says nothing about price, latency, or how a model behaves on your own cases, which is where Vitna's engine measures quality.
Source: OpenRouter (openrouter.ai/models), as of 17 Nov 16:29 UTC. sha256 0bb53a77713f over 853,270 bytes.
Catalog: OpenRouter's public model list.
Intelligence, coding and agentic indexes: Artificial Analysis, as published by OpenRouter.
Elo, win rate and rank: Design Arena, as published by OpenRouter.
Vitna ran none of these evaluations and adjusted none of these figures.
