Model comparison
GPT-4 vs Olmo 2 0325 32b Instruct
Olmo 2 0325 32b Instruct is the stronger model overall, scoring 32.7 to 29.1 on the Noometry Index.
Last verified . 11 shared benchmarks.
Allen Institute for AI (Ai2)Summary
- They share 11 benchmarks with published results for both. GPT-4 scores higher in 3 categories and Olmo 2 0325 32b Instruct in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Olmo 2 0325 32b Instruct leads 26.8 to 10.8.
- Olmo 2 0325 32b Instruct has downloadable open weights; the other is API-only.
Side by side
| GPT-4 | Olmo 2 0325 32b Instruct | |
|---|---|---|
| Provider | OpenAI | Allen Institute for AI (Ai2) |
| Noometry Index | 29.1 | 32.7 |
| Released | 2023-03-14 | — |
| Weights | Proprietary | Open |
| Context window | 8K | — |
| Max output | 8K | — |
| Input $ / M tokens | $30 | — |
| Output $ / M tokens | $60 | — |
| Results tracked | 38 | 16 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Olmo 2 0325 32b Instruct leads
GPT-4: 31.6 (#283), Olmo 2 0325 32b Instruct: 35.2 (#227)
| Benchmark | GPT-4 | Olmo 2 0325 32b Instruct |
|---|---|---|
| LMArena Coding | 1254 | 1210 |
| WeirdML | 12.4% | — |
| BigCodeBench Instruct | 46% | — |
| BigCodeBench Complete | 57.2% | — |
| HumanEval+ | 79.3% | — |
Agentic & Tool Use Not comparable
GPT-4: —, Olmo 2 0325 32b Instruct: —
| Benchmark | GPT-4 | Olmo 2 0325 32b Instruct |
|---|---|---|
| METR Time Horizons | 36.1% | — |
Reasoning Olmo 2 0325 32b Instruct leads
GPT-4: 17.8 (#289), Olmo 2 0325 32b Instruct: 23.6 (#175)
| Benchmark | GPT-4 | Olmo 2 0325 32b Instruct |
|---|---|---|
| LMArena Hard Prompts | 1241 | 1208 |
| Chess Puzzles | 4% | — |
| Mystery Game Puzzles | 12% | — |
| DTBench | 62.7% | — |
| LMCA | 17.1% | — |
| BIG-Bench Hard | 75.1% | — |
| Epoch Capabilities Index | 125.89 | — |
| ForecastBench | 57.8 | — |
| HellaSwag | 95.3% | — |
| WinoGrande | 87.5% | — |
Math Olmo 2 0325 32b Instruct leads
GPT-4: 10.8 (#309), Olmo 2 0325 32b Instruct: 26.8 (#255)
| Benchmark | GPT-4 | Olmo 2 0325 32b Instruct |
|---|---|---|
| LMArena Math | 1269 | 1208 |
| OTIS Mock AIME 2024-2025 | 1.1% | — |
| Omni-MATH | — | 16.1% |
| MATH Level 5 | 23% | — |
| GSM8K | 92% | — |
Knowledge Olmo 2 0325 32b Instruct leads
GPT-4: 18.4 (#282), Olmo 2 0325 32b Instruct: 19.5 (#279)
| Benchmark | GPT-4 | Olmo 2 0325 32b Instruct |
|---|---|---|
| GPQA Diamond | 35.7% | — |
| MMLU-Pro | — | 41.4% |
| GPQA (HELM) | — | 28.7% |
| LMArena Expert | 1211 | — |
| MMLU | 86.4% | — |
| TriviaQA | 84.8% | — |
Multilingual GPT-4 leads
GPT-4: 40.6 (#215), Olmo 2 0325 32b Instruct: 34.8 (#248)
| Benchmark | GPT-4 | Olmo 2 0325 32b Instruct |
|---|---|---|
| LMArena Non-English | 1246 | 1160 |
| LMArena Chinese | 1242 | 1192 |
| LMArena Russian | 1251 | 1187 |
| LMArena French | 1283 | — |
| LMArena German | 1251 | — |
| LMArena Japanese | 1209 | — |
| LMArena Korean | 1184 | — |
| LMArena Spanish | 1261 | — |
Instruction Following GPT-4 leads
GPT-4: 65.3 (#222), Olmo 2 0325 32b Instruct: 61.5 (#244)
| Benchmark | GPT-4 | Olmo 2 0325 32b Instruct |
|---|---|---|
| LMArena Instruction Following | 1241 | 1186 |
| IFEval | — | 78% |
Long Context GPT-4 leads
GPT-4: 37.7 (#212), Olmo 2 0325 32b Instruct: 36.2 (#234)
| Benchmark | GPT-4 | Olmo 2 0325 32b Instruct |
|---|---|---|
| LMArena Longer Query | 1244 | 1194 |
Writing & Preference Olmo 2 0325 32b Instruct leads
GPT-4: 34.9 (#268), Olmo 2 0325 32b Instruct: 42.1 (#236)
| Benchmark | GPT-4 | Olmo 2 0325 32b Instruct |
|---|---|---|
| LMArena Text | 1263 | 1218 |
| LMArena Creative Writing | 1244 | 1199 |
| LMArena Multi-Turn | 1257 | 1221 |
| EQ-Bench Creative Writing | 752 | — |
| WildBench | — | 73.4% |
Frequently asked questions
Is GPT-4 better than Olmo 2 0325 32b Instruct?
Olmo 2 0325 32b Instruct is the stronger model overall, scoring 32.7 to 29.1 on the Noometry Index.
Is GPT-4 or Olmo 2 0325 32b Instruct better for coding?
Olmo 2 0325 32b Instruct scores higher on coding benchmarks: 35.2 versus 31.6 in the Noometry coding category.
How many benchmarks do GPT-4 and Olmo 2 0325 32b Instruct share?
11 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and Olmo 2 0325 32b Instruct has 16.