Olmo 3 32B Think places in the top three on 6 benchmarks.
Ai2 Nov 2025 Open weights Flagship
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| AGIEval | 88.2% | #2 of 40 | Lab model card | |
| BIG-Bench Hard | 89.8% | #6 of 73 | Lab model card | |
| AlpacaEval 2.0 | 74.2% | #2 of 16 | Lab model card | |
| IFEval | 89% | #45 of 135 | Lab model card | |
| MMLU | 85.4% | #53 of 131 | Lab model card | |
| PopQA | 31.9% | #3 of 7 | Lab model card | |
| CulturalRiddles | 11.3% | #12 of 20 | Official leaderboard | |
| LMArena (Elo) | 1,306 | #224 of 335 | Official leaderboard | |
| IFBench | 47.6% | #66 of 80 | Lab model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MATH | 96.1% | #3 of 99 | Lab model card | |
| AIME 2024 | 76.8% | #49 of 88 | Lab model card | |
| GPQA Diamond | 58.1% | #225 of 302 | Lab model card | |
| AIME 2025 | 72.5% | #107 of 143 | Lab model card | |
| ZebraLogic | 76% | #20 of 23 | Lab model card | |
| OMEGA | 50.8% | #2 of 2 | Lab model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LiveCodeBench | 83.5% | #6 of 54 | Lab model card | |
| HumanEval+ | 91.4% | #8 of 27 | Lab model card | |
| MBPP+ | 68% | #13 of 15 | Lab model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LisanBench | 333.7 | #78 of 133 | Official leaderboard | |
| SnakeBench | 19 | #142 of 188 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| BBQ (Bias Benchmark for QA) | 88.2% | #9 of 25 | Lab model card | |
| WildGuardTest | 89.4% | #2 of 5 | Lab model card | |
| HarmBench | 73.5% | #3 of 5 | Lab model card | |
| WMDP | 34.8% | #2 of 3 | Lab model card | |
| XSTest | 93.9% | #12 of 16 | Lab model card | |
| StrongREJECT | 80.8% | #15 of 19 | Lab model card | |
| Tulu 3 Safety Eval | 68.8% | #6 of 6 | Lab model card | |
| DoAnythingNow (DAN) | 20.2% | #2 of 2 | Lab model card | |
| TrustLLM JailbreakTrigger | 73.3% | #2 of 2 | Lab model card |
The Self-Reported Index fits one score to the 14 capability results Ai2 reported for this model, on the scale of the Epoch Capabilities Index. Labs choose which benchmarks to show and how to run them, so read it as the lab's claim, not an independent measurement. The 90% interval runs from 131.0 to 136.8. How we compute it.
The benchmarks the lab chose to headline, with the scores it gave.
Every result in the longer documents the lab published with it. 10 of them aren't in the launch post.