Home / Best AI for… / Best AI for Math
Task ranking
Best AI for Math
Ranked by MMLU-Pro — the only benchmark we track with dedicated mathematics subject categories — across every model it has measured.
Our top pick: Qwen3.8-Max — MMLU-Pro 88.6% · GPQA Diamond 92.6% · League Score 62 · $2.00/$6.00 per 1M. Full record, pricing and benchmarks on its model page; or run the model picker for a recommendation matched to your budget and context needs.
| # | Model | Maker | Why it ranks here |
|---|---|---|---|
| 1 | Qwen3.8-Max | Alibaba Qwen | MMLU-Pro 88.6% · GPQA Diamond 92.6% · League Score 62 · $2.00/$6.00 per 1M |
| 2 | GPT-5 | OpenAI | MMLU-Pro 86.5% · League Score 60 · $1.25/$10.00 per 1M |
| 3 | GPT-5.6 Luna | OpenAI | MMLU-Pro 86.0% · GPQA Diamond 92.3% · League Score 60 · $0.20/$1.20 per 1M |
| 4 | Grok 4.3 | xAI (SpaceXAI) | MMLU-Pro 85.8% · GPQA Diamond 90.1% · League Score 74 · $1.25/$2.50 per 1M |
| 5 | Gemini 3.5 Flash-Lite | Google DeepMind | MMLU-Pro 85.8% · League Score 70 · $0.30/$2.50 per 1M |
| 6 | Nemotron 3 Ultra 550B-A55B | NVIDIA | MMLU-Pro 85.8% · GPQA Diamond 87.0% · League Score 57 |
| 7 | DeepSeek V4 Pro (0813) | DeepSeek | MMLU-Pro 82.9% · GPQA Diamond 90.1% · League Score 63 · $1.32/$3.96 per 1M |
| 8 | Nemotron 3 Nano 30B-A3B | NVIDIA | MMLU-Pro 78.3% · GPQA Diamond 73.0% · League Score 0 |
Rankings regenerate from the model database on every data update. Metrics and sources: methodology.