LLM leaderboard
Every tracked language model by intelligence, with what it costs and how fast it answers.
ranked by intelligence · 611 models · updated Aug 30 · 51–100
| # | Model | Creator | Intelligence | $/1M in | $/1M out | tok/s |
|---|---|---|---|---|---|---|
| 51 | Motif 3 | Motif Technologies | 47.40 | — | — | — |
| 52 | GPT-5.6 Luna (high) | OpenAI | 47.00 | $0.2 | $1.2 | 118 |
| 53 | GPT-5.6 Terra (medium) | OpenAI | 46.80 | $2 | $12 | 96 |
| 54 | Gemini 3.5 Flash (medium) | 46.70 | $1.5 | $9 | 207 | |
| 55 | Qwen3.7 Max | Alibaba | 46.70 | $2.5 | $7.5 | 199 |
| 56 | GPT-5.3 Codex (xhigh) | OpenAI | 45.50 | $1.75 | $14 | 122 |
| 57 | MiniMax-M3 | MiniMax | 45.40 | $0.3 | $1.2 | 148 |
| 58 | Motif 3 (Beta) | Motif Technologies | 45.30 | — | — | — |
| 59 | DeepSeek V4 Pro (Reasoning, Max Effort) | DeepSeek | 45.30 | $0.43 | $0.87 | 66 |
| 60 | Kimi K2.6 | Kimi | 45.10 | $0.95 | $4 | 38 |
| 61 | Claude Opus 4.6 (Adaptive Reasoning, Max Effort) | Anthropic | 44.90 | $5 | $25 | 44 |
| 62 | Qwen3.8 27B (medium) | Alibaba | 44.50 | $0.5 | $3 | 52 |
| 63 | GPT-5.5 (low) | OpenAI | 44.50 | $5 | $30 | 74 |
| 64 | Muse Spark | Meta | 44.30 | — | — | — |
| 65 | Claude Opus 4.7 (Non-reasoning, High Effort) | Anthropic | 43.90 | $5 | $25 | 41 |
| 66 | DeepSeek V4 Pro (Reasoning, High Effort) | DeepSeek | 43.70 | $0.43 | $0.87 | 63 |
| 67 | GPT-5.2 (xhigh) | OpenAI | 43.30 | $1.75 | $14 | 75 |
| 68 | Kimi K2.7 Code | Kimi | 43.00 | $0.95 | $4 | 46 |
| 69 | MiMo-V2.5-Pro | Xiaomi | 42.90 | $0.43 | $0.87 | 33 |
| 70 | Qwen3.8 27B (low) | Alibaba | 42.90 | $0.5 | $3 | 57 |
| 71 | Claude Sonnet 5 (Non-reasoning, High Effort) | Anthropic | 42.60 | $2 | $10 | 64 |
| 72 | Inkling (xhigh) | Thinking Machines | 42.30 | $1 | $4.05 | 75 |
| 73 | Hy3 | Tencent | 42.20 | $0.14 | $0.55 | 67 |
| 74 | DeepSeek V4 Flash (Reasoning, Max Effort) | DeepSeek | 42.10 | $0.13 | $0.28 | — |
| 75 | Claude Opus 4.5 (Reasoning) | Anthropic | 41.90 | $5 | $25 | 46 |
| 76 | GPT-5.6 Sol (Non-reasoning) | OpenAI | 41.90 | $4 | $20 | 68 |
| 77 | Nex-N2-Pro | Nex AGI | 41.70 | $0.5 | $2.5 | 130 |
| 78 | Solar Pro 4 | Upstage | 41.60 | $0.3 | $1.2 | 44 |
| 79 | MiMo-V2-Pro | Xiaomi | 41.40 | — | — | — |
| 80 | GPT-5.6 Terra (low) | OpenAI | 41.30 | $2 | $12 | 102 |
| 81 | GPT-5.2 Codex (xhigh) | OpenAI | 41.20 | $1.75 | $14 | — |
| 82 | Inkling Small | Thinking Machines | 41.20 | $0.3 | $1.2 | 76 |
| 83 | Qwen3.6 Max Preview | Alibaba | 41.10 | $1.3 | $7.8 | 35 |
| 84 | GLM-5.1 (Reasoning) | Z AI | 41.00 | $1.38 | $4.4 | 61 |
| 85 | GPT-5.4 mini (xhigh) | OpenAI | 40.90 | $0.75 | $4.5 | 164 |
| 86 | Grok Build 0.1 0616 | SpaceXAI | 40.70 | $1 | $2 | 55 |
| 87 | GLM-5 (Reasoning) | Z AI | 40.60 | $1 | $3.2 | 71 |
| 88 | Gemini 3 Pro Preview (high) | 40.60 | $2 | $12 | — | |
| 89 | Qwen3.6 Plus | Alibaba | 40.50 | $0.5 | $3 | 55 |
| 90 | GPT-5.4 (low) | OpenAI | 40.20 | $2.5 | $15 | 103 |
| 91 | JT-4.1 Flash 236B A21B | China Mobile | 39.90 | — | — | — |
| 92 | Agnes 2.5 Pro Alpha | Sapiens AI | 39.70 | $0.45 | $0.9 | 130 |
| 93 | GPT-5.4 nano (xhigh) | OpenAI | 39.70 | $0.2 | $1.25 | 163 |
| 94 | Qwen3.7 Plus | Alibaba | 39.40 | $0.4 | $1.6 | 56 |
| 95 | GLM-5-Turbo | Z AI | 39.10 | — | — | — |
| 96 | DeepSeek V4 Flash (Reasoning, High Effort) | DeepSeek | 39.00 | $0.13 | $0.28 | — |
| 97 | MiniMax-M2.7 | MiniMax | 38.90 | $0.3 | $1.2 | 66 |
| 98 | GPT-5.6 Luna (medium) | OpenAI | 38.90 | $0.2 | $1.2 | 114 |
| 99 | GPT-5.2 (medium) | OpenAI | 38.90 | $1.75 | $14 | — |
| 100 | Claude Opus 4.6 (Non-reasoning, High Effort) | Anthropic | 38.80 | $5 | $25 | 36 |
what intelligence means
A composite of nine independent evaluations covering reasoning, coding, agentic work and knowledge. Higher is better. It is versioned, so scores are comparable within a version rather than across all time.