Which AI model scores highest on AIME 2024?

gpt-oss-120b by OpenAI holds the best AIME 2024 result among tracked models, at 95.8% (released Aug 5 2025). Higher scores are better on this benchmark.

What are the top 5 models on AIME 2024?

1. gpt-oss-120b (OpenAI) — 95.8%; 2. gpt-oss-20b (OpenAI) — 92.1%; 3. Gemini 2.5 Pro (Google) — 92%; 4. o1-mini (OpenAI) — 70%; 5. GPT-4.1 mini (OpenAI) — 49.6%.

How many models have a published AIME 2024 score?

9 tracked models have a published AIME 2024 score. Scores are collected from published lab reports and benchmark sources, with recorded sources credited alongside the results. We do not re-run the evaluations.

What is the best open model on AIME 2024?

gpt-oss-120b by OpenAI is the highest-ranked model with downloadable weights on AIME 2024, scoring 95.8% at rank 1 overall.

Reasoning

AIME 2024

single-answer accuracy on the 2024 invitational mathematics exam without tools

Rankings

Higher is better

AIME 2024 — frequently asked questions

: single-answer accuracy on the 2024 invitational mathematics exam without tools
: gpt-oss-120b by OpenAI holds the best AIME 2024 result among tracked models, at 95.8% (released Aug 5 2025). Higher scores are better on this benchmark.
: 1. gpt-oss-120b (OpenAI) — 95.8%; 2. gpt-oss-20b (OpenAI) — 92.1%; 3. Gemini 2.5 Pro (Google) — 92%; 4. o1-mini (OpenAI) — 70%; 5. GPT-4.1 mini (OpenAI) — 49.6%.
: 9 tracked models have a published AIME 2024 score. Scores are collected from published lab reports and benchmark sources, with recorded sources credited alongside the results. We do not re-run the evaluations.
: gpt-oss-120b by OpenAI is the highest-ranked model with downloadable weights on AIME 2024, scoring 95.8% at rank 1 overall.