Reported by others
Math proofs (ProofBench)
Proof problems at the level of math competitions and graduate courses; the model writes a proof and it is graded for correctness.
How to read this table
- Higher rows are rated higher. A range like “1–2” means the gap is within the margin of error.
- The score is the value published by the task's authors and other organizations.
- Conditions can differ between models, so small gaps do not mean much.
- Prices are a guide per million tokens, reading then writing.
- Length is how much text the model can read at once (tokens).
| Rank | Model | Developer | Accuracy (%) | Input $ /1M | Output $ /1M | Context | Supports | Weights | Released | Price and specs |
|---|---|---|---|---|---|---|---|---|---|---|
| 1–5 | Claude Fable 5.1 | Anthropic | Rating 100.0 | $10 | $50 | 1M | Anthropic Released 2026-09 Price $10 / $50 Length 1M | |||
| 1–5 | Claude Opus 5.5 | Anthropic | Rating 100.0 | $4 | $20 | 1M | Anthropic Released 2026-09 Price $4 / $20 Length 1M | |||
| 1–5 | Claude Sonnet 5.5 | Anthropic | Rating 100.0 | $2 | $10 | 1M | Anthropic Released 2026-09 Price $2 / $10 Length 1M | |||
| 1–9 | Claude Opus 5 | Anthropic | Rating 99.0 ±8.2 | $5 | $25 | 1M | Anthropic Released 2026-07 Price $5 / $25 Length 1M | |||
| 4–8 | Gemini 4 Argon | Rating 99.0 | $2 | $10 | 1.05M | Google Released 2026-09 Price $2 / $10 Length 1.05M | ||||
| 4–8 | GPT-6.1 Sol | OpenAI | Rating 99.0 | $2 | $10 | 1.05M | OpenAI Released 2026-09 Price $2 / $10 Length 1.05M | |||
| 4–8 | GPT-6 Astra | OpenAI | Rating 99.0 | $10 | $50 | 1.05M | OpenAI Released 2026-09 Price $10 / $50 Length 1.05M | |||
| 1–10 | Claude Fable 5 | Anthropic | Rating 95.0 ±8.3 | $10 | $50 | 1M | Anthropic Released 2026-06 Price $10 / $50 Length 1M | |||
| 8–11 | Claude Haiku 5.5 | Anthropic | Rating 89.0 | $0.1 | $0.5 | 1M | Anthropic Released 2026-10 Price $0.1 / $0.5 Length 1M | |||
| 7–14 | Kimi K3 | Moonshot | Rating 87.0 ±9.0 | $3 | $15 | 1.05M | Moonshot Released 2026-07 Price $3 / $15 Length 1.05M | |||
| 10–13 | Muse Wonder | Meta | Rating 84.0 | $4 | $20 | 1.05M | Meta Released 2026-10 Price $4 / $20 Length 1.05M | |||
| 9–16 | GPT-5.6 Sol | OpenAI | Rating 83.0 ±8.2 | $4 | $20 | 1.05M | OpenAI Released 2026-07 Price $4 / $20 Length 1.05M | |||
| 11–14 | GPT-6 Sol | OpenAI | Rating 83.0 | $2 | $10 | 1.05M | OpenAI Released 2026-09 Price $2 / $10 Length 1.05M | |||
| 10–18 | Claude Sonnet 5 | Anthropic | Rating 77.0 ±9.3 | $2 | $10 | 1M | Anthropic Released 2026-06 Price $2 / $10 Length 1M | |||
| 13–17 | Tencent HY4 (preview) | Tencent | Rating 75.0 | — | — | — | Tencent Released 2026-08 Price — Length — | |||
| 13–18 | GPT-5.6 Terra | OpenAI | Rating 74.0 ±8.9 | $2 | $12 | 1.05M | OpenAI Released 2026-07 Price $2 / $12 Length 1.05M | |||
| 15–18 | MiMo-V2.6 Pro | Xiaomi | Rating 70.0 | $0.435 | $0.87 | 1.05M | Xiaomi Released 2026-09 Price $0.435 / $0.87 Length 1.05M | |||
| 14–22 | Claude Opus 4.8 | Anthropic | Rating 69.0 ±9.1 | $5 | $25 | 1M | Anthropic Released 2026-05 Price $5 / $25 Length 1M | |||
| 18–21 | GPT-6 Luna | OpenAI | Rating 64.0 | $0.1 | $0.5 | 1.05M | OpenAI Released 2026-09 Price $0.1 / $0.5 Length 1.05M | |||
| 19–23 | MiMo-V2.6 Flash | Xiaomi | Rating 63.0 | $0.14 | $0.28 | 1.05M | Xiaomi Released 2026-09 Price $0.14 / $0.28 Length 1.05M | |||
| 18–31 | GPT-5.6 Luna | OpenAI | Rating 60.0 ±9.8 | $0.2 | $1.2 | 1.05M | OpenAI Released 2026-07 Price $0.2 / $1.2 Length 1.05M | |||
| 21–28 | Gemini 3.7 Flash | Rating 58.0 | $0.75 | $3.75 | 1.05M | Google Released 2026-08 Price $0.75 / $3.75 Length 1.05M | ||||
| 21–28 | Muse Spark 1.3 | Meta | Rating 58.0 | $1.25 | $4.25 | 1.05M | Meta Released 2026-09 Price $1.25 / $4.25 Length 1.05M | |||
| 21–28 | Qwen3.8 Max | Alibaba | Rating 58.0 | $2 | $6 | 1M | Alibaba Released 2026-08 Price $2 / $6 Length 1M | |||
| 24–29 | DeepSeek-V4 Flash (2026-07) | DeepSeek | Rating 56.0 | — | — | 1M | DeepSeek Released 2026-07 Price — Length 1M | |||
| 18–35 | GPT-5.4 | OpenAI | Rating 56.0 ±9.7 | $2.5 | $15 | 1.05M | OpenAI Released 2026-03 Price $2.5 / $15 Length 1.05M | |||
| 20–35 | Claude Opus 4.7 | Anthropic | Rating 54.0 ±9.8 | $5 | $25 | 1M | Anthropic Released 2026-04 Price $5 / $25 Length 1M | |||
| 25–31 | DeepSeek-V4.1 Flash | DeepSeek | Rating 54.0 | — | — | 1M | DeepSeek Released 2026-09 Price — Length 1M | |||
| 26–32 | Grok 4.6 | xAI | Rating 51.0 | $2 | $6 | 500K | xAI Released 2026-08 Price $2 / $6 Length 500K | |||
| 21–38 | Claude Opus 4.6 | Anthropic | Rating 50.0 ±9.8 | $5 | $25 | 1M | Anthropic Released 2026-02 Price $5 / $25 Length 1M | |||
| 28–33 | DeepSeek-V4 Pro (2026-08) | DeepSeek | Rating 50.0 | $0.66 | $1.98 | 1M | DeepSeek Released 2026-08 Price $0.66 / $1.98 Length 1M | |||
| 21–38 | GPT-5.5 | OpenAI | Rating 50.0 ±9.8 | $5 | $30 | 1.05M | OpenAI Released 2026-04 Price $5 / $30 Length 1.05M | |||
| 29–34 | GLM-5.3 | Z.ai | Rating 49.0 | $1.4 | $4.4 | 1M | Z.ai Released 2026-08 Price $1.4 / $4.4 Length 1M | |||
| 30–36 | Gemini 3.8 Flash | Rating 48.0 | $0.75 | $3.75 | 1.05M | Google Released 2026-09 Price $0.75 / $3.75 Length 1.05M | ||||
| 26–41 | Claude Sonnet 4.6 | Anthropic | Rating 45.0 ±9.8 | $3 | $15 | 1M | Anthropic Released 2026-02 Price $3 / $15 Length 1M | |||
| 33–40 | Muse Spark 1.2 | Meta | Rating 43.0 | $1.25 | $4.25 | 1.05M | Meta Released 2026-08 Price $1.25 / $4.25 Length 1.05M | |||
| 34–41 | Step 5 (preview) | StepFun | Rating 42.0 | $1.0461 | $2.989 | 1M | StepFun Released 2026-09 Price $1.0461 / $2.989 Length 1M | |||
| 32–43 | Muse Spark 1.1 | Meta | Rating 39.0 ±9.6 | $1.25 | $4.25 | 1M | Meta Released 2026-07 Price $1.25 / $4.25 Length 1M | |||
| 35–45 | Claude Opus 4.5 | Anthropic | Rating 36.0 ±9.4 | $5 | $25 | 200K | Anthropic Released 2025-11 Price $5 / $25 Length 200K | |||
| 35–45 | Gemini 3.6 Flash | Rating 36.0 ±9.5 | $0.75 | $3.75 | 1.05M | Google Released 2026-07 Price $0.75 / $3.75 Length 1.05M | ||||
| 35–46 | GLM-5.2 | Z.ai | Rating 35.0 ±9.4 | $1.4 | $4.4 | 1M | Z.ai Released 2026-06 Price $1.4 / $4.4 Length 1M | |||
| 38–50 | Gemini 3.5 Flash | Rating 31.0 ±8.9 | $1.5 | $9 | 1.05M | Google Released 2026-05 Price $1.5 / $9 Length 1.05M | ||||
| 38–50 | Grok 4.5 | xAI | Rating 31.0 ±9.0 | $2 | $6 | 500K | xAI Released 2026-07 Price $2 / $6 Length 500K | |||
| 39–60 | Gemini 3.1 Pro (preview) | Rating 26.0 ±8.6 | $2 | $12 | 1.05M | Google Released 2026-02 Price $2 / $12 Length 1.05M | ||||
| 41–51 | Grok 4.7 | xAI | Rating 26.0 | $2 | $6 | 500K | xAI Released 2026-09 Price $2 / $6 Length 500K | |||
| 39–60 | Qwen3.7 Max | Alibaba | Rating 26.0 ±8.6 | $2.5 | $7.5 | 1M | Alibaba Released 2026-05 Price $2.5 / $7.5 Length 1M | |||
| 42–64 | GLM-5.1 | Z.ai | Rating 22.2 ±8.2 | $1.4 | $4.4 | 200K | Z.ai Released 2026-04 Price $1.4 / $4.4 Length 200K | |||
| 42–64 | MiMo-V2.5 Pro | Xiaomi | Rating 22.0 ±8.4 | $0.435 | $0.87 | 1.05M | Xiaomi Released 2026-04 Price $0.435 / $0.87 Length 1.05M | |||
| 45–60 | GLM-5.3 Flash | Z.ai | Rating 21.0 | $0.15 | $0.5 | 1M | Z.ai Released 2026-08 Price $0.15 / $0.5 Length 1M | |||
| 42–65 | GPT-5.4 mini | OpenAI | Rating 21.0 ±8.0 | $0.75 | $4.5 | 400K | OpenAI Released 2026-03 Price $0.75 / $4.5 Length 400K | |||
| 42–65 | Gemini 3 Pro (preview) | Rating 20.0 ±7.8 | — | — | 1M | Google Released 2025-11 Price — Length 1M | ||||
| 44–65 | Claude Sonnet 4.5 | Anthropic | Rating 19.0 ±7.7 | $3 | $15 | 200K | Anthropic Released 2025-09 Price $3 / $15 Length 200K | |||
| 45–68 | GPT-5 | OpenAI | Rating 18.0 ±7.5 | $1.25 | $10 | 400K | OpenAI Released 2025-08 Price $1.25 / $10 Length 400K | |||
| 45–68 | MiniMax-M3 | MiniMax | Rating 18.0 ±7.7 | $0.3 | $1.2 | 1.05M | MiniMax Released 2026-06 Price $0.3 / $1.2 Length 1.05M | |||
| 45–70 | Muse Spark | Meta | Rating 17.0 ±7.4 | — | — | — | Meta Released 2026-04 Price — Length — | |||
| 45–68 | DeepSeek-V4 Pro | DeepSeek | Rating 16.0 ±5.9 | $0.66 | $1.98 | 1M | DeepSeek Released 2026-04 Price $0.66 / $1.98 Length 1M | |||
| 45–70 | Kimi K2.6 | Moonshot | Rating 16.0 ±7.2 | $0.95 | $4 | 262.1K | Moonshot Released 2026-04 Price $0.95 / $4 Length 262.1K | |||
| 45–70 | MiMo-V2.5 | Xiaomi | Rating 16.0 ±7.2 | $0.14 | $0.28 | 1.05M | Xiaomi Released 2026-04 Price $0.14 / $0.28 Length 1.05M | |||
| 48–65 | Qwen3.8 27B | Alibaba | Rating 16.0 | — | — | 262.1K | Alibaba Released 2026-08 Price — Length 262.1K | |||
| 45–70 | Gemini 3 Flash (preview) | Rating 15.0 ±7.0 | $0.5 | $3 | 1M | Google Released 2025-12 Price $0.5 / $3 Length 1M | ||||
| 45–70 | GPT-5.2 | OpenAI | Rating 15.0 ±7.0 | $1.75 | $14 | 400K | OpenAI Released 2025-12 Price $1.75 / $14 Length 400K | |||
| 48–72 | Grok 4.20 Reasoning | xAI | Rating 14.0 ±6.8 | $1.25 | $2.5 | 2M | xAI Released 2026-02 Price $1.25 / $2.5 Length 2M | |||
| 48–72 | Gemini 3.5 Flash-Lite | Rating 13.0 ±6.6 | $0.3 | $2.5 | 1.05M | Google Released 2026-07 Price $0.3 / $2.5 Length 1.05M | ||||
| 48–73 | GPT-5 nano | OpenAI | Rating 12.0 ±6.4 | $0.05 | $0.4 | 400K | OpenAI Released 2025-08 Price $0.05 / $0.4 Length 400K | |||
| 50–75 | Grok 4.3 | xAI | Rating 11.0 ±6.1 | $1.25 | $2.5 | 1M | xAI Released 2026-04 Price $1.25 / $2.5 Length 1M | |||
| 57–72 | Mistral Large 4 | Mistral | Rating 10.0 | $1.36 | $4.18 | 524.3K | Mistral Released 2026-10 Price $1.36 / $4.18 Length 524.3K | |||
| 54–76 | GPT-5.1 Codex | OpenAI | Rating 9.0 ±5.6 | — | — | 400K | OpenAI Released 2025-11 Price — Length 400K | |||
| 54–76 | GPT-5 mini | OpenAI | Rating 9.0 ±5.6 | $0.25 | $2 | 400K | OpenAI Released 2025-08 Price $0.25 / $2 Length 400K | |||
| 54–76 | Mistral Medium 3.5 | Mistral | Rating 9.0 ±5.9 | $1.5 | $7.5 | 262.1K | Mistral Released 2026-04 Price $1.5 / $7.5 Length 262.1K | |||
| 57–77 | DeepSeek-V3.2 | DeepSeek | Rating 8.0 ±5.3 | — | — | 128K | DeepSeek Released 2025-12 Price — Length 128K | |||
| 62–78 | GLM-4.7 | Z.ai | Rating 6.0 ±4.7 | $0.6 | $2.2 | 204.8K | Z.ai Released 2025-12 Price $0.6 / $2.2 Length 204.8K | |||
| 62–78 | Inkling Small | Thinking Machines | Rating 6.0 ±4.3 | $0.58 | $1.44 | 524.3K | Thinking Machines Released 2026-07 Price $0.58 / $1.44 Length 524.3K | |||
| 65–79 | GPT-5.4 nano | OpenAI | Rating 5.0 ±4.3 | $0.2 | $1.25 | 400K | OpenAI Released 2026-03 Price $0.2 / $1.25 Length 400K | |||
| 66–79 | Grok 4.1 Fast Reasoning | xAI | Rating 4.0 ±3.8 | — | — | 2M | xAI Released 2025-11 Price — Length 2M | |||
| 66–79 | MiniMax-M2.5 | MiniMax | Rating 4.0 ±3.8 | $0.3 | $1.2 | 204.8K | MiniMax Released 2026-02 Price $0.3 / $1.2 Length 204.8K | |||
| 70–78 | Mercury 2.5 | Inception | Rating 3.0 | — | — | — | Inception Released 2026-09 Price — Length — | |||
| 67–81 | MiniMax-M2.7 | MiniMax | Rating 3.0 ±3.3 | $0.3 | $1.2 | 204.8K | MiniMax Released 2026-03 Price $0.3 / $1.2 Length 204.8K | |||
| 71–81 | NVIDIA Nemotron 3 Ultra | NVIDIA | Rating 2.0 ±2.8 | — | — | 131.1K | NVIDIA Released 2026-06 Price — Length 131.1K | |||
| 74–81 | Inkling | Thinking Machines | Rating 0.0 ±2.8 | $1.87 | $4.68 | 262.1K | Thinking Machines Released 2026-07 Price $1.87 / $4.68 Length 262.1K | |||
| 77–81 | Laguna M.1 | Poolside | Rating 0.0 ±0.0 | — | — | 262.1K | Poolside Released 2026-04 Price — Length 262.1K | |||
| 77–81 | Laguna XS.2 | Poolside | Rating 0.0 ±0.0 | — | — | 32.8K | Poolside Released 2026-04 Price — Length 32.8K |
Best accuracy over time
2025-08-07Release date2026-10-12
How to read this page
- The value is the share of proofs judged correct. Figures published by the organizations that made the tasks, among others, collected here.
- The accuracy is the share of problems solved. A gap of 1 point is 1 question out of 100.
- In the Supports column the icons are, from left: can use tools / can read images / can answer in a fixed format (JSON); faded means not supported or unknown.
- Open-weights models have their contents (the weights) published, so you can also run them on your own computer or server. Models without the open-weights mark are normally used through the developer's own services (apps or API); for some models the weights status is unknown.
- Names such as ChatGPT next to a model are the consumer services its developer offers. Check on the service itself whether that model is available there.
Data: Arena, Epoch AI, Swallow LLM team (CC BY 4.0), Models.dev (MIT), updated 2026-10-24, data sources