Best AI for mathematics
As of 2026-07-18, Anthropic: Claude Opus 4.8 (Fast) is the strongest model for this task in our daily benchmark, where over 350 AI models are evaluated and the best are presented.
Language models do not calculate — they predict text. That they nevertheless solve maths well comes from having learned to reason through the problem step by step. It is also why they can be confidently wrong.
For maths we rank by overall hvilkenAI score, the single best available measure of reasoning ability. Ranked by overall score, from today’s benchmark.
Today’s ranking
| Model | Provider | Score | Language | Speed | Input price/1M |
|---|---|---|---|---|---|
| Anthropic: Claude Opus 4.8 (Fast) | anthropic | 8.3/10 | 9.3/10 | 131 t/s | $10.00 |
| Anthropic: Claude Haiku 4.5 | anthropic | 8.0/10 | 9.3/10 | 113 t/s | $1.00 |
| Google: Gemma 3 4B | 8.0/10 | 9.3/10 | 237 t/s | $0.05 | |
| Mistral: Mistral Small 3 | mistralai | 7.8/10 | 8.7/10 | 234 t/s | $0.05 |
| Perplexity: Sonar Pro | perplexity | 7.7/10 | 9.0/10 | 68 t/s | $3.00 |
| Cohere: Command R+ (08-2024) | cohere | 7.5/10 | 9.3/10 | 160 t/s | $2.50 |
| Meta: Llama 4 Maverick | meta-llama | 7.2/10 | 8.0/10 | 176 t/s | $0.20 |
| Google: Gemini 3.5 Flash | 7.2/10 | 8.0/10 | 419 t/s | $1.50 |
Ask for the working, not just the answer
Ask for a number and you get a number — with no way to tell whether it is right. Ask for every step and you get a chain of reasoning you can check, and models make fewer errors when forced to work step by step.
Always verify
The dangerous error is the one that looks right: correct method, correct setup, wrong arithmetic in step three. Check the final answer with a calculator or spreadsheet when it matters, and ask the model to re-check its own work.
Frequently asked questions
Other tasks
Doing something else? See which model is best for it.
Read also