AI model leaderboard
All 42 models we rank, and where each one places for every kind of task. Updated Nov 2, 2026.
Sort by
| # | Model | Coding | Writing | Math | Research | Everyday chat | API price |
|---|---|---|---|---|---|---|---|
| 1 | Claude Opus 5.5 | #1 | #1 | #1 | #1 | #1 | $4 / $20 |
| 2 | Gemini 4 Argon | #2 | #3 | #2 | #2 | #2 | $2 / $10 |
| 3 | Claude Fable 5.1 | #3 | #2 | #3 | #2 | #3 | $10 / $50 |
| 4 | Claude Sonnet 5.5 | #4 | #4 | #4 | #4 | #4 | $2 / $10 |
| 5 | MiMo V2.6 Pro | #5 | #10 | #8 | #10 | #11 | $0.43 / $0.87 |
| 6 | GPT-6 Astra | #6 | #9 | #6 | #8 | #5 | $10 / $50 |
| 7 | Muse Spark 1.3 | #7 | #8 | #7 | #9 | #9 | $1.25 / $4.25 |
| 8 | GPT-6.1 Sol | #8 | #12 | #5 | #30 | #13 | $2 / $10 |
| 9 | ChatGPT 5.6 Sol | #9 | #13 | #12 | #22 | #16 | $4 / $20 |
| 10 | Qwen 3.8 Max | #10 | #5 | #10 | #18 | #8 | $2 / $6 |
| 11 | Kimi K3 | #11 | #11 | #13 | #6 | #10 | $3 / $15 |
| 12 | GLM 5.3 | #12 | #6 | #11 | #7 | #7 | $1.40 / $4.40 |
| 13 | Gemini 3.8 Flash | #13 | #17 | #9 | #11 | #6 | $0.75 / $3.75 |
| 14 | GLM 5.3 Flash | #14 | #7 | #14 | #5 | #12 | $0.15 / $0.50 |
| 15 | DeepSeek V4.1 Flash 1017 | #15 | #14 | #15 | #13 | #17 | $0.30 / $1.20 |
| 16 | Claude Haiku 5.5 | #16 | #16 | #16 | #14 | #19 | $0.10 / $0.50 |
| 17 | DeepSeek V4.1 Flash | #17 | #20 | #19 | #15 | #21 | $0.30 / $1.20 |
| 18 | Ling 3.1 Flash | #18 | #19 | #17 | #17 | #20 | $0.30 / $0.90 |
| 19 | Muse Spark Open | #19 | #21 | #20 | #19 | #22 | $1.25 / $4.25 |
| 20 | ChatGPT 5.5 | #20 | #15 | #18 | #26 | #14 | $5 / $30 |
| 21 | Mistral Large 4 | #21 | #22 | #21 | #20 | #23 | $0.70 / $2.10 |
| 22 | DeepSeek V4 Pro | #22 | #18 | #24 | #12 | #18 | $1.32 / $3.96 |
| 23 | GPT 5.3 Codex | #23 | #26 | #22 | #24 | #26 | $1.75 / $14 |
| 24 | Gemini 3.1 Pro Preview | #24 | #24 | #23 | #21 | #25 | $2 / $12 |
| 25 | DeepSeek V4 Flash | #25 | #28 | #28 | #25 | #27 | $0.13 / $0.26 |
| 26 | GPT-6 Luna | #26 | #32 | #27 | #40 | #28 | $0.10 / $0.50 |
| 27 | Kimi K2.6 | #27 | #27 | #25 | #16 | #24 | $0.95 / $4 |
| 28 | NVIDIA Nemotron 3.5 Super VL | #28 | #31 | #26 | #33 | #29 | $0.20 / $0.50 |
| 29 | MiniMax M3 | #29 | #34 | #36 | #31 | #33 | $0.30 / $1.20 |
| 30 | Qwen 3.7 Plus | #30 | #33 | #31 | #32 | #34 | $0.40 / $1.60 |
| 31 | GPT 5.5 Pro | #31 | #29 | #32 | #27 | #30 | $30 / $180 |
| 31 | Grok 4.20 Reasoning | #31 | #29 | #32 | #27 | #30 | $1.25 / $2.50 |
| 31 | Muse Wonder | #31 | #23 | #32 | #27 | #30 | $4 / $20 |
| 31 | Qwen 3.8 Flash | #31 | #25 | #32 | #23 | #15 | $0.15 / $0.47 |
| 35 | Tencent Hy3 | #35 | #35 | #30 | #35 | #36 | $0.14 / $0.58 |
| 36 | Inkling | #36 | #38 | #29 | #34 | #35 | $1 / $4.05 |
| 37 | GLM 5V Turbo | #37 | #36 | #38 | #37 | #38 | $1.20 / $4 |
| 38 | Nemotron 3 Ultra | #38 | #39 | #37 | #36 | #37 | $0.60 / $2.40 |
| 39 | Gemini 3.5 Flash Lite | #39 | #37 | #39 | #38 | #39 | $0.30 / $2.50 |
| 40 | Gemma 4 31B IT | #40 | #40 | #40 | #39 | #40 | $0.14 / $0.40 |
| 41 | Claude Haiku 4.5 | #41 | #41 | #41 | #41 | #41 | $1 / $5 |
| 42 | GPT OSS 120B | #42 | #42 | #42 | #42 | #42 | $0.10 / $0.50 |
A rank is about the quality of the answers, blending public leaderboards with our own blind tests where we have them. API price is per 1M tokens, in / out, and never moves a rank. How we rank
The best for each task: coding, writing, math, research, everyday chat.