Mistral Small 3.2 24B Instruct 2506 — Benchmarks
Benchmark scores for Mistral Small 3.2 24B Instruct 2506 aggregated from public leaderboards, with how it ranks among open models. See hardware requirements for what you need to run it.
Overall rank: #69 of 78 open modelscomposite 24.3/100 across 6 benchmarks in 3 categories · methodology
Coding
| Benchmark | Score | Open rank | All models |
|---|---|---|---|
| SciCode | 26.4% | #53 / 62 | #150 / 161 |
Instruction Following
| Benchmark | Score | Open rank | All models |
|---|---|---|---|
| LMArena Text | 1356.7 | #66 / 182 | #187 / 391 |
Math
| Benchmark | Score | Open rank | All models |
|---|---|---|---|
| AIME 2024/2025 | 30.3% | #47 / 82 | #202 / 270 |
Reasoning
| Benchmark | Score | Open rank | All models |
|---|---|---|---|
| Chess Puzzles | 1.0% | #44 / 65 | #180 / 205 |
| DTBench | 59.9% | #48 / 67 | #168 / 210 |
| GPQA Diamond | 49.0% | #53 / 94 | #216 / 289 |
| CritPt | 0.0% | #57 / 62 | #163 / 172 |
Scores aggregated from public benchmark sources (each linked from the benchmark pages). llmrun does not run these benchmarks.