Division
Value Division
League Score per blended dollar — capability per dollar, the table no one else publishes.
| # | Model | Maker | Division metric | League Score |
|---|---|---|---|---|
| 1 | Claude Haiku 5.5 | Anthropic | 339.5 pts per $ · $0.20 blended | 68 |
| 2 | GPT-6 Luna | OpenAI | 278.5 pts per $ · $0.20 blended | 56 |
| 3 | DeepSeek V4.1 Flash | DeepSeek | 143.6 pts per $ · $0.52 blended | 75 |
| 4 | GPT-5.6 Luna | OpenAI | 134.0 pts per $ · $0.45 blended | 60 |
| 5 | Gemini 3.5 Flash-Lite | Google DeepMind | 82.2 pts per $ · $0.85 blended | 70 |
| 6 | Grok 4.3 | xAI (SpaceXAI) | 47.2 pts per $ · $1.56 blended | 74 |
| 7 | Gemini 3.8 Flash | Google DeepMind | 42.0 pts per $ · $1.50 blended | 63 |
| 8 | Kimi K2.6 | Moonshot AI | 40.7 pts per $ · $1.71 blended | 70 |
| 9 | Muse Spark 1.3 | Meta | 34.4 pts per $ · $2.00 blended | 69 |
| 10 | DeepSeek V4 Pro (0813) | DeepSeek | 31.9 pts per $ · $1.98 blended | 63 |
| 11 | GPT-6.1 Sol | OpenAI | 23.4 pts per $ · $4.00 blended | 94 |
| 12 | Gemini 4 Argon | Google DeepMind | 22.7 pts per $ · $4.00 blended | 91 |
| 13 | Grok 4.7 | xAI (SpaceXAI) | 22.3 pts per $ · $3.00 blended | 67 |
| 14 | Claude Sonnet 5.5 | Anthropic | 21.7 pts per $ · $4.00 blended | 87 |
| 15 | Qwen3.8-Max | Alibaba Qwen | 20.6 pts per $ · $3.00 blended | 62 |
| 16 | Mistral Medium 3.5 | Mistral AI | 19.7 pts per $ · $3.00 blended | 59 |
| 17 | Claude Sonnet 5 | Anthropic | 18.5 pts per $ · $4.00 blended | 74 |
| 18 | GPT-5 | OpenAI | 17.5 pts per $ · $3.44 blended | 60 |
| 19 | GPT-5.6 Terra | OpenAI | 16.4 pts per $ · $4.50 blended | 74 |
| 20 | Claude Opus 5.5 | Anthropic | 12.1 pts per $ · $8.00 blended | 96 |
| 21 | Gemini 3.1 Pro Preview | Google DeepMind | 11.7 pts per $ · $4.50 blended | 52 |
| 22 | Kimi K3 | Moonshot AI | 10.5 pts per $ · $6.00 blended | 63 |
| 23 | GPT-5.6 Sol | OpenAI | 10.3 pts per $ · $8.00 blended | 82 |
| 24 | Muse Wonder | Meta | 9.6 pts per $ · $8.00 blended | 76 |
| 25 | Claude Opus 5 | Anthropic | 9.2 pts per $ · $10.00 blended | 92 |
| 26 | Claude Haiku 4.5 | Anthropic | 8.2 pts per $ · $2.00 blended | 16 |
| 27 | Claude Fable 5.1 | Anthropic | 4.9 pts per $ · $20.00 blended | 97 |
| 28 | GPT-6 Astra | OpenAI | 4.8 pts per $ · $20.00 blended | 97 |
| 29 | Mistral Large 4 | Mistral AI | 3.9 pts per $ · $1.03 blended | 4 |
Division tables regenerate from the model database on every update. Scoring rules: methodology.