winner on MATH-500
OpenAI: o3 Mini97.9
direct benchmark result, not a broad vertical composite | source row dated 2000-01-01
scored on 2000-01-01 · stale source data (9646d)
latest mapped results | top 20
| # | Model | Score | Evidence | Tested |
|---|---|---|---|---|
| 1 | OpenAI: o3 Mini | 97.9 | model-only independent_benchmark | 2000-01-01 |
| 2 | OpenAI: o1 | 96.4 | model-only independent_benchmark | 2000-01-01 |
| 3 | Google: Gemini 2.0 Flash | 89.7 | model-only independent_benchmark | 2000-01-01 |
| 4 | OpenAI: GPT-5 | 84.7 | model-only independent_benchmark | 2000-01-01 |
| 5 | Meta: Llama 3.3 70B Instruct | 77.0 | model-only independent_benchmark | 2000-01-01 |
| 6 | Qwen: Qwen3 235B A22B | 71.8 | model-only independent_benchmark | 2000-01-01 |
| 7 | Meta: Llama 4 Maverick | 61.2 | model-only independent_benchmark | 2000-01-01 |
| 8 | Qwen2.5 Coder 32B Instruct | 57.2 | model-only independent_benchmark | 2000-01-01 |
what this result means
MATH-500 benchmark scores from LLM-Stats.
This benchmark contributes direct public evidence. Read its scope before generalizing the result.
A win here is a win on MATH-500. Broad task pages require independent corroboration before naming a general winner.
source record
category: math
metric: accuracy
matched models: 8
latest source date: 2000-01-01
direction: higher is better