Intelligence Index
Artificial Analysis's Intelligence Index: a composite of the publisher's own evaluation runs across reasoning, knowledge and related tasks, on a 0 to 100 scale, as OpenRouter carries it. 58 models ranked, as carried in OpenRouter's catalog.
retrieved sha256 · 0bb53a77713fartificialanalysis.ai ↗
The top 26 of 58, 64.6 down to 29.6 on the Intelligence Index. 6 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.
6 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 3.1 Flash29.6 · $0.068 blended
- 02Claude Haiku 5.536.5 · $0.20 blended · +6.9 at 3.0x
- 03GPT-6.1 Terra51.3 · $2.00 blended · +14.8 at 10x
- 04Gemini 4 Hydrogen53.9 · $3.00 blended · +2.6 at 1.5x
- 05Gemini 4 Carbon60.0 · $4.00 blended · +6.1 at 1.3x
- 06Claude Fable 5.264.6 · $16.00 blended · +4.6 at 4.0x
57 of 58 ranked models carry a list price; 1 is 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. 8 of 57 priced models sit in the attractive quadrant: at or above the median score of 28.5 and at or below the median blended price of $1.05. The tinted corner is that quadrant.
All 58, best first.
Artificial Analysis · 0 to 1009 more authors are on this board, the largest with 2 models. Search a name to reach one.
| Position | Model | Author | Score | In / out per M | Context | Listed |
|---|---|---|---|---|---|---|
| 01 | Claude Fable 5.2 | Anthropic | 64.6 | $8.00 / $40.00 | 1M | |
| 02 | Gemini 4 Carbon | 60.0 | $2.00 / $10.00 | 1.05M | ||
| 03 | Claude Opus 5.5 | Anthropic | 59.1 | $4.00 / $20.00 | 1M | |
| 04 | GPT-6 Astra | OpenAI | 56.9 | $10.00 / $50.00 | 1.05M | |
| 05 | GPT-6.1 Sol | OpenAI | 54.6 | $2.00 / $10.00 | 1.05M | |
| 06 | Claude Sonnet 5.5 | Anthropic | 54.5 | $2.00 / $10.00 | 1M | |
| 07 | Gemini 4 Hydrogen | 53.9 | $1.50 / $7.50 | 1.05M | ||
| 08 | GPT-6.1 Terra | OpenAI | 51.3 | $1.00 / $5.00 | 1.05M | |
| 09 | Gemini 4 Argon | 50.4 | $2.00 / $10.00 | 1.05M | ||
| 10 | Claude Fable 5.1 | Anthropic | 50.0 | $10.00 / $50.00 | 1M | |
| 11 | Kimi K3.1 Pro | MoonshotAI | 46.5 | $0.24 / $8.40 | 1.05M | |
| 12 | Claude Opus 5 | Anthropic | 44.2 | $5.00 / $25.00 | 1M | |
| 13 | GPT-6 Sol | OpenAI | 44.2 | $2.00 / $10.00 | 1.05M | |
| 14 | Muse Wonder | Meta | 43.6 | $4.00 / $20.00 | 1.05M | |
| 15 | Muse Spark 1.3 | Meta | 36.6 | $1.25 / $4.25 | 1.05M | |
| 16 | Claude Haiku 5.5 | Anthropic | 36.5 | $0.10 / $0.50 | 1M | |
| 17 | Grok 4.7 | SpaceXAI | 35.5 | $2.00 / $6.00 | 500k | |
| 18 | Hy4 | Tencent | 34.6 | $0.417 / $1.25 | 1.05M | |
| 19 | Qwen3.8 Max (0902) | Qwen | 33.7 | $2.00 / $6.00 | 1M | |
| 20 | GPT-5.6 Terra | OpenAI | 32.3 | $2.00 / $12.00 | 1.05M | |
| 21 | GLM 5.3 | Z.ai | 31.7 | $0.0699 / $1.40 | 1.05M | |
| 22 | Grok 4.6 | SpaceXAI | 31.0 | $2.00 / $6.00 | 500k | |
| 23 | Gemini 3.8 Flash | 30.8 | $0.825 / $4.13 | 1.05M | ||
| 24 | Inkling 2 Small | Thinking Machines | 30.6 | $0.30 / $1.00 | 1.05M | |
| 25 | MiMo-V2.6-Pro | Xiaomi | 29.9 | $0.435 / $0.87 | 1.05M | |
| 26 | Ling 3.1 Flash | inclusionAI | 29.6 | $0.045 / $0.135 | 1M | |
| 27 | MiniMax M3.1 Flash | MiniMax | 29.1 | $0.075 / $0.30 | 1.05M | |
| 28 | DeepSeek V4.1 Flash 1017 | DeepSeek | 28.6 | $0.036 / $0.50 | 1.05M | |
| 29 | Step 5 Preview | StepFun | 28.5 | $1.00 / $2.70 | 1M | |
| 30 | Solar Pro 4 | Upstage | 28.2 | $0.15 / $0.60 | 524k | |
| 31 | GPT-6 Luna | OpenAI | 27.9 | $0.10 / $0.50 | 1.05M | |
| 32 | GLM 5.3 Flash | Z.ai | 27.8 | $0.15 / $0.50 | 1.05M | |
| 33 | MiMo-V2.6-Flash | Xiaomi | 26.7 | $0.14 / $0.28 | 1.05M | |
| 34 | Mistral Large 4 | Mistral | 26.1 | $0.70 / $2.10 | 1.05M | |
| 35 | Muse Spark Open | Meta | 24.9 | $1.25 / $4.25 | 1.05M | |
| 36 | GPT-5.6 Luna | OpenAI | 23.8 | $0.20 / $1.20 | 1.05M | |
| 37 | DeepSeek V4 Pro 0813 | DeepSeek | 23.1 | $0.111 / $4.44 | 1.05M | |
| 38 | DeepSeek V4 Pro 0423 | DeepSeek | 20.8 | $0.90 / $1.50 | 1.05M | |
| 39 | Qwen3.8 27B | Qwen | 20.1 | $0.15 / $1.88 | 1.02M | |
| 40 | MiniMax M3 | MiniMax | 18.9 | $0.30 / $1.20 | 1.05M | |
| 41 | Beam 1.0 501B A23B | Reflection | 17.7 | $0.30 / $1.00 | 1.05M | |
| 42 | Kimi K3 | MoonshotAI | 17.4 | $0.192 / $16.00 | 1.05M | |
| 43 | Nemotron 3.5 Super VL | NVIDIA | 13.6 | $0.15 / $0.50 | 1.05M | |
| 44 | Inkling Small | Thinking Machines | 12.9 | $0.45 / $1.20 | 524k | |
| 45 | Inkling | Thinking Machines | 12.5 | $1.00 / $4.05 | 524k | |
| 46 | Nemotron 3 Ultra | NVIDIA | 11.8 | $0.50 / $2.20 | 262k | |
| 47 | Gemini 3.5 Flash Lite | 11.7 | $0.30 / $2.50 | 1.05M | ||
| 48 | Claude Haiku 4.5 | Anthropic | 7.4 | $1.00 / $5.00 | 200k | |
| 49 | Muse Glimmer 30B | Meta | 6.3 | $0.35 / $1.50 | 131k | |
| 50 | Command A+ | Cohere | 6.2 | $0.30 / $1.50 | 192k | |
| 51 | Gemini 3.1 Flash Lite Preview | 6.2 | $0.25 / $1.50 | 1.05M | ||
| 52 | Nemotron 3.5 Lightning | NVIDIA | 5.5 | $0.07 / $0.20 | 1.05M | |
| 53 | Granite 4.2 8B | IBM Granite | 5.1 | $0.06 / $0.15 | 131k | |
| 54 | Mistral Medium 3.5 | Mistral | 4.6 | $1.50 / $7.50 | 262k | |
| 55 | Gemma 4 31B | 4.4 | $0.20 / $0.40 | 262k | ||
| 56 | Mistral Small 4 | Mistral | 4.3 | $0.15 / $0.60 | 262k | |
| 57 | gpt-oss-120b | OpenAI | 3.8 | $0.15 / $0.60 | 131k | |
| 58 | North Mini Code (free) | Cohere | 3.4 | free | 256k |
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.
