Model comparison
GPT-4 vs Olmo 3.1 32b Think
Olmo 3.1 32b Think is the stronger model overall, scoring 37.3 to 28.9 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-4 scores higher in 1 category and Olmo 3.1 32b Think in 7 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in math, where Olmo 3.1 32b Think leads 33.6 to 10.1.
- Olmo 3.1 32b Think has downloadable open weights; the other is API-only.
Side by side
| GPT-4 | Olmo 3.1 32b Think | |
|---|---|---|
| Provider | OpenAI | Allen Institute for AI (Ai2) |
| Noometry Index | 28.9 | 37.3 |
| Released | 2023-03-14 | — |
| Weights | Proprietary | Open |
| Context window | 8K | — |
| Max output | 8K | — |
| Input $ / M tokens | $30 | — |
| Output $ / M tokens | $60 | — |
| Results tracked | 38 | 15 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Olmo 3.1 32b Think leads
GPT-4: 31.2 (#277), Olmo 3.1 32b Think: 37.1 (#186)
| Benchmark | GPT-4 | Olmo 3.1 32b Think |
|---|---|---|
| LMArena Coding | 1254 | 1292 |
| WeirdML | 12.4% | — |
| BigCodeBench Instruct | 46% | — |
| BigCodeBench Complete | 57.2% | — |
| HumanEval+ | 79.3% | — |
Agentic & Tool Use Not comparable
GPT-4: —, Olmo 3.1 32b Think: —
| Benchmark | GPT-4 | Olmo 3.1 32b Think |
|---|---|---|
| METR Time Horizons | 36.1% | — |
Reasoning Olmo 3.1 32b Think leads
GPT-4: 17.3 (#285), Olmo 3.1 32b Think: 24.1 (#151)
| Benchmark | GPT-4 | Olmo 3.1 32b Think |
|---|---|---|
| LMArena Hard Prompts | 1241 | 1272 |
| Chess Puzzles | 4% | — |
| Mystery Game Puzzles | 12% | — |
| DTBench | 62.7% | — |
| LMCA | 17.1% | — |
| BIG-Bench Hard | 75.1% | — |
| Epoch Capabilities Index | 125.88 | — |
| ForecastBench | 57.8 | — |
| HellaSwag | 95.3% | — |
| WinoGrande | 87.5% | — |
Math Olmo 3.1 32b Think leads
GPT-4: 10.1 (#304), Olmo 3.1 32b Think: 33.6 (#170)
| Benchmark | GPT-4 | Olmo 3.1 32b Think |
|---|---|---|
| LMArena Math | 1269 | 1305 |
| OTIS Mock AIME 2024-2025 | 1.1% | — |
| MATH Level 5 | 23% | — |
| GSM8K | 92% | — |
Knowledge Olmo 3.1 32b Think leads
GPT-4: 18.4 (#277), Olmo 3.1 32b Think: 35.0 (#178)
| Benchmark | GPT-4 | Olmo 3.1 32b Think |
|---|---|---|
| LMArena Expert | 1211 | 1294 |
| GPQA Diamond | 35.7% | — |
| MMLU | 86.4% | — |
| TriviaQA | 84.8% | — |
Multilingual GPT-4 leads
GPT-4: 40.8 (#210), Olmo 3.1 32b Think: 38.3 (#226)
| Benchmark | GPT-4 | Olmo 3.1 32b Think |
|---|---|---|
| LMArena Non-English | 1246 | 1210 |
| LMArena Chinese | 1242 | 1241 |
| LMArena French | 1282 | 1262 |
| LMArena German | 1251 | 1260 |
| LMArena Russian | 1251 | 1193 |
| LMArena Spanish | 1261 | 1288 |
| LMArena Japanese | 1209 | — |
| LMArena Korean | 1184 | — |
Instruction Following Too close to call
GPT-4: 65.5 (#217), Olmo 3.1 32b Think: 65.8 (#213)
| Benchmark | GPT-4 | Olmo 3.1 32b Think |
|---|---|---|
| LMArena Instruction Following | 1241 | 1246 |
Long Context Too close to call
GPT-4: 37.7 (#207), Olmo 3.1 32b Think: 38.6 (#190)
| Benchmark | GPT-4 | Olmo 3.1 32b Think |
|---|---|---|
| LMArena Longer Query | 1244 | 1272 |
Writing & Preference Olmo 3.1 32b Think leads
GPT-4: 35.0 (#263), Olmo 3.1 32b Think: 46.4 (#215)
| Benchmark | GPT-4 | Olmo 3.1 32b Think |
|---|---|---|
| LMArena Text | 1263 | 1271 |
| LMArena Creative Writing | 1244 | 1226 |
| LMArena Multi-Turn | 1257 | 1252 |
| EQ-Bench Creative Writing | 752 | — |
Frequently asked questions
Is GPT-4 better than Olmo 3.1 32b Think?
Olmo 3.1 32b Think is the stronger model overall, scoring 37.3 to 28.9 on the Noometry Index.
Is GPT-4 or Olmo 3.1 32b Think better for coding?
Olmo 3.1 32b Think scores higher on coding benchmarks: 37.1 versus 31.2 in the Noometry coding category.
How many benchmarks do GPT-4 and Olmo 3.1 32b Think share?
15 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and Olmo 3.1 32b Think has 15.