Model comparison
GPT-4o vs Olmo 3.1 32b Think
Olmo 3.1 32b Think is the stronger model overall, scoring 37.9 to 28.6 on the Noometry Index.
Last verified . 15 shared benchmarks.
Allen Institute for AI (Ai2)Summary
- They share 15 benchmarks with published results for both. GPT-4o scores higher in 4 categories and Olmo 3.1 32b Think in 4 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in math, where Olmo 3.1 32b Think leads 36.3 to 10.6.
- Olmo 3.1 32b Think has downloadable open weights; the other is API-only.
Side by side
| GPT-4o | Olmo 3.1 32b Think | |
|---|---|---|
| Provider | OpenAI | Allen Institute for AI (Ai2) |
| Noometry Index | 28.6 | 37.9 |
| Released | 2024-05-13 | — |
| Weights | Proprietary | Open |
| Context window | 128K | — |
| Max output | 16K | — |
| Input $ / M tokens | $2.50 | — |
| Output $ / M tokens | $10 | — |
| Results tracked | 72 | 15 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Olmo 3.1 32b Think leads
GPT-4o: 24.8 (#328), Olmo 3.1 32b Think: 37.7 (#189)
| Benchmark | GPT-4o | Olmo 3.1 32b Think |
|---|---|---|
| LMArena Coding | 1297 | 1291 |
| SWE-bench Verified | 31% | — |
| SWE-bench Verified (bash only) | 21.6% | — |
| Aider Polyglot | 45.3% | — |
| GSO | 0% | — |
| WeirdML | 25.1% | — |
| BigCodeBench Instruct | 51.1% | — |
| LiveBench Coding | 51.4% | — |
| BigCodeBench Complete | 61.1% | — |
| CadEval | 26% | — |
| HumanEval+ | 87.2% | — |
| MBPP+ | 72.2% | — |
Agentic & Tool Use Not comparable
GPT-4o: 21.0 (#141), Olmo 3.1 32b Think: —
| Benchmark | GPT-4o | Olmo 3.1 32b Think |
|---|---|---|
| GDPval | 9.9% | — |
| TheAgentCompany | 8.6% | — |
| Cybench | 12.5% | — |
| BALROG | 32.3% | — |
| LMArena Search | 1006 | — |
| METR Time Horizons | 40.8% | — |
Reasoning Olmo 3.1 32b Think leads
GPT-4o: 9.4 (#343), Olmo 3.1 32b Think: 25.2 (#150)
| Benchmark | GPT-4o | Olmo 3.1 32b Think |
|---|---|---|
| LMArena Hard Prompts | 1281 | 1272 |
| ARC-AGI-2 | 0% | — |
| SimpleBench | 17.8% | — |
| ARC-AGI-1 | 4.5% | — |
| CritPt | 0% | — |
| Chess Puzzles | 13% | — |
| EnigmaEval | 0.8% | — |
| LiveBench Reasoning | 55.8% | — |
| DTBench | 64.5% | — |
| LiveBench Data Analysis | 60.9% | — |
| LMCA | 16.6% | — |
| Epoch Capabilities Index | 128.97 | — |
| ForecastBench | 57.7 | — |
| LiveBench | 55.3% | — |
Math Olmo 3.1 32b Think leads
GPT-4o: 10.6 (#312), Olmo 3.1 32b Think: 36.3 (#168)
| Benchmark | GPT-4o | Olmo 3.1 32b Think |
|---|---|---|
| LMArena Math | 1285 | 1305 |
| FrontierMath (Tiers 1-3) | 0.4% | — |
| OTIS Mock AIME 2024-2025 | 6.4% | — |
| Omni-MATH | 29.3% | — |
| LiveBench Math | 49.5% | — |
| MATH Level 5 | 53.3% | — |
| FrontierMath (Feb 2025 set) | 0.3% | — |
Knowledge Olmo 3.1 32b Think leads
GPT-4o: 28.8 (#242), Olmo 3.1 32b Think: 35.7 (#181)
| Benchmark | GPT-4o | Olmo 3.1 32b Think |
|---|---|---|
| LMArena Expert | 1250 | 1295 |
| GPQA Diamond | 49.2% | — |
| Humanity's Last Exam | 2.7% | — |
| SimpleQA Verified | 26% | — |
| MMLU-Pro | 71.3% | — |
| Confabulations | 15.3% | — |
| Vectara Hallucination Rate | 9.6% | — |
| GPQA (HELM) | 52% | — |
| MMLU | 88.1% | — |
Multimodal Not comparable
GPT-4o: 34.5 (#91), Olmo 3.1 32b Think: —
| Benchmark | GPT-4o | Olmo 3.1 32b Think |
|---|---|---|
| LMArena Vision | 1137 | — |
| Video-MME | 71.9% | — |
| GeoBench | 71% | — |
| VPCT | 40% | — |
| ScienceQA | 88.5% | — |
Multilingual GPT-4o leads
GPT-4o: 43.2 (#186), Olmo 3.1 32b Think: 38.1 (#231)
| Benchmark | GPT-4o | Olmo 3.1 32b Think |
|---|---|---|
| LMArena Non-English | 1283 | 1209 |
| LMArena Chinese | 1277 | 1242 |
| LMArena French | 1304 | 1260 |
| LMArena German | 1282 | 1262 |
| LMArena Russian | 1286 | 1193 |
| LMArena Spanish | 1292 | 1289 |
| LMArena Japanese | 1257 | — |
| LMArena Korean | 1234 | — |
Instruction Following Too close to call
GPT-4o: 66.6 (#207), Olmo 3.1 32b Think: 65.6 (#218)
| Benchmark | GPT-4o | Olmo 3.1 32b Think |
|---|---|---|
| LMArena Instruction Following | 1278 | 1247 |
| LiveBench Instruction Following | 68.6% | — |
| IFEval | 81.7% | — |
Long Context Too close to call
GPT-4o: 39.4 (#179), Olmo 3.1 32b Think: 38.6 (#195)
| Benchmark | GPT-4o | Olmo 3.1 32b Think |
|---|---|---|
| LMArena Longer Query | 1289 | 1272 |
| Fiction.LiveBench | 66.7% | — |
Writing & Preference GPT-4o leads
GPT-4o: 52.6 (#166), Olmo 3.1 32b Think: 46.2 (#220)
| Benchmark | GPT-4o | Olmo 3.1 32b Think |
|---|---|---|
| LMArena Text | 1300 | 1272 |
| LMArena Creative Writing | 1292 | 1226 |
| LMArena Multi-Turn | 1302 | 1252 |
| Short-Story Creative Writing | 81.8% | — |
| WildBench | 82.8% | — |
| LiveBench Language | 47.6% | — |
Frequently asked questions
Is GPT-4o better than Olmo 3.1 32b Think?
Olmo 3.1 32b Think is the stronger model overall, scoring 37.9 to 28.6 on the Noometry Index.
Is GPT-4o or Olmo 3.1 32b Think better for coding?
Olmo 3.1 32b Think scores higher on coding benchmarks: 37.7 versus 24.8 in the Noometry coding category.
How many benchmarks do GPT-4o and Olmo 3.1 32b Think share?
15 benchmarks have published results for both models. GPT-4o has 72 scored results on Noometry and Olmo 3.1 32b Think has 15.