Normie ModeBest AI for maths and science
Maths problems, science questions and engineering.
Get it with Claude Max, $100 a month (for up to half your weekly allowance)
Top of 12 models for this task
Get it with Google AI Pro, $19.99 a month
2nd of 12 models for this task
3rd and 5th of 12 models for this task
Based on tests of this task. The picks follow a fixed rule using the charts below and each app's plans.
Norm's guide
What I'd tell a friend who asked.
What AI is good at
- Working through problems step by step.
- Explaining the idea behind a formula.
- Checking your answer.
Where it trips up
- Small slips in long calculations.
- Harder problems, where it may guess confidently.
- Units and rounding.
How to get a good result
- Ask it to show every step.
- If the answer matters, check it another way.
- For calculations, ask it to use a calculator or code if it can.
How the models compare
Every model we could score for maths and science, combining the tests below.
Overall for maths and science
Position on a combined score from 5 tests. The longer the bar, the further ahead · Higher is better
On the tests we have, Claude Fable 5.1 comes out on top, followed by Gemini 3.8 Flash and Kimi K3.
The evidence
Competition maths
Tests this task · Problems in the style of a top US high-school maths competition. · Higher is better
Claude Fable 5.1, GPT-5.6 Sol and GPT-6 Astra are level at the top.
Research-level maths
Tests this task · New, unpublished maths problems that take specialists hours or days. · Higher is better
GPT-6 Astra and Claude Fable 5.1 are neck and neck at the top, followed by GPT-5.6 Sol and GPT-5.6 Luna.
Graduate-level science questions
Tests this task · Biology, physics and chemistry questions written so they can't be answered by searching the web. · Higher is better
GPT-6 Astra and Gemini 3.8 Flash are neck and neck at the top, followed by Gemini 3.1 Pro and Gemini 3.6 Flash; GPT-6 Astra costs about 13 times as much.
People's votes on maths questions
Tests this task · People compared two anonymous answers to maths questions. · Higher is better
Gemini 3.8 Flash and Claude Fable 5.1 are neck and neck at the top, followed by Gemini 3.6 Flash and Kimi K3; Claude Fable 5.1 costs about 13 times as much.
People's votes on science questions
Tests a related skill · People compared two anonymous answers to questions in the life, physical and social sciences. · Higher is better
Claude Opus 5.5 and Claude Fable 5.1 are neck and neck at the top, followed by Gemini 3.8 Flash and Kimi K3; Claude Fable 5.1 costs about 3 times as much.
If you build with it
What the makers charge developers who use the models directly. Using an app? The plans above are what you pay.
Cost per 1,000 typical requests
List price, about 1,500 words in and 500 out per request · Lower is better
GPT-6 Luna is cheapest, followed by DeepSeek V4.1 Flash and GPT-5.6 Luna.