Skip to content
← All frontiers

🧮 Math / Reasoning Tutor frontier

As of October 25, 2026, the highest Math / Reasoning Tutor score on this frontier is Claude Opus 5.5 at 100.0, with an estimated monthly cost of $198.00.

Tutors students through hard math. Reasoning-heavy; cost matters less because depth-of-thought is the value. Weighted toward contamination-resistant problem sets (FrontierMath, OTIS Mock AIME) since AIME 2024 and MATH-500 are now in most training corpora.

The y axis is this scenario's weighted score; the x axis starts at estimated monthly cost for 100 requests a day. Switch axes to compare against speed, latency or context. See the full ranking.

On the frontier

Models on the Pareto frontier, highest score first
ModelScoreEst. $/month
Claude Opus 5.5Anthropic100.0$198
GPT-6.1 SolOpenAI99.7$50
DeepSeek V4.1 Flash 1017DeepSeek99.4$6
Claude Haiku 5.5Anthropic93.2$5
GPT-6 LunaOpenAI91.6$2
DeepSeek V4 Flash 0731DeepSeek87.3$2
Qwen3.7 FlashAlibaba (Qwen)75.2$1
gpt-oss-20bOpenAI56.4$1
Gemma 3 4BGoogle2.7$1