Leaderboard

Best AI for Math

Ranked by AIME 2025 — competition-level math problems requiring multi-step reasoning.

11 models ranked · Updated April 2026

1

Qwen solves 85% of AIME 2025 problems — top of the consumer-AI math leaderboard.

# Model AIME 2025 Standard plan Context
1 Qwen Alibaba · Qwen 3 Max 85% Plus 256K Compare →
2 ChatGPT OpenAI · GPT-5.2 84% $20/mo 256K Compare →
3 Microsoft Copilot Microsoft · GPT-5.2 (M365-tuned) 82% $20/mo 128K Compare →
4 Claude Opus Anthropic · Claude Opus 4.7 81% $100/mo 1M Compare →
5 Gemini Google · Gemini 3 Pro 80% $20/mo 2M Compare →
6 Claude Anthropic · Claude Sonnet 4.6 78% $20/mo 1M Compare →
7 Grok xAI · Grok 4 75% $30/mo 256K Compare →
8 DeepSeek DeepSeek · DeepSeek V3 72% API only 128K Compare →
9 Meta AI Meta · Llama 4 Maverick 70% Free 1M Compare →
10 Le Chat Mistral · Mistral Large 2 65% $14.99/mo 128K Compare →
11 Pi Inflection AI · Pi 3 55% Free 32K Compare →
Not benchmarked on this metric
Methodology

AIME 2025 — American Invitational Mathematics Examination, multi-step reasoning. Self-reported by vendors and corroborated by lmarena.ai.

Or skip the choice

One subscription. Every frontier AI model. $14.99/month.

Perspective AI bundles ChatGPT-class, Claude, Gemini, Grok, Copilot and more in one app. Switch mid-conversation. No per-vendor logins, no separate bills.

Launch app →