Model comparison
GPT-4.5 vs gpt-oss-120b
GPT-4.5 and gpt-oss-120b score almost the same on the Noometry Index (36.9 vs 36.0), so choose on price, context window or the category you care about most.
Last verified . 26 shared benchmarks.
Summary
- They share 26 benchmarks with published results for both. GPT-4.5 scores higher in 6 categories and gpt-oss-120b in 3 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where gpt-oss-120b leads 50.6 to 31.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 37.8% for GPT-4.5 and 88.9% for gpt-oss-120b.
- gpt-oss-120b has downloadable open weights; the other is API-only.
Side by side
| GPT-4.5 | gpt-oss-120b | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 36.9 | 36.0 |
| Released | 2025-02-27 | 2025-08-05 |
| Weights | Proprietary | Open |
| Context window | — | 131K |
| Max output | — | 41K |
| Input $ / M tokens | — | $0.037 |
| Output $ / M tokens | — | $0.17 |
| Results tracked | 42 | 48 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-4.5 leads
GPT-4.5: 41.8 (#102), gpt-oss-120b: 33.2 (#251)
| Benchmark | GPT-4.5 | gpt-oss-120b |
|---|---|---|
| Aider Polyglot | 44.9% | 41.8% |
| WeirdML | 39.4% | 48.2% |
| LMArena Coding | 1397 | 1380 |
| SWE-bench Verified (bash only) | — | 26% |
| SciCode | — | 36% |
| LiveBench Coding | 75.2% | — |
| ALE-Bench | — | 575.62 |
| AlgoTune | — | 1.41 |
Agentic & Tool Use GPT-4.5 leads
GPT-4.5: 27.6 (#92), gpt-oss-120b: 12.1 (#145)
| Benchmark | GPT-4.5 | gpt-oss-120b |
|---|---|---|
| Terminal-Bench | — | 18.7% |
| APEX-Agents | — | 4.4% |
| Cybench | 17.5% | — |
| METR Time Horizons | — | 56.6% |
| Vending-Bench 2 | — | -21.53 |
Reasoning gpt-oss-120b leads
GPT-4.5: 14.1 (#320), gpt-oss-120b: 19.5 (#236)
| Benchmark | GPT-4.5 | gpt-oss-120b |
|---|---|---|
| SimpleBench | 34.5% | 22.1% |
| LMArena Hard Prompts | 1404 | 1364 |
| Epoch Capabilities Index | 136.74 | 139.94 |
| ARC-AGI-2 | 0.8% | — |
| Kagi LLM Benchmark | — | 58.6% |
| ARC-AGI-1 | 10.3% | — |
| CritPt | — | 1.1% |
| Chess Puzzles | — | 20% |
| EnigmaEval | 3.2% | — |
| LiveBench Reasoning | 71.1% | — |
| Mystery Game Puzzles | — | 2% |
| DTBench | — | 76.3% |
| LiveBench Data Analysis | 64.3% | — |
| LMCA | — | 22.1% |
| Surface Evolver Bench | — | 25% |
| ForecastBench | 61.7 | — |
| LiveBench | 69% | — |
Math gpt-oss-120b leads
GPT-4.5: 31.0 (#200), gpt-oss-120b: 50.6 (#43)
| Benchmark | GPT-4.5 | gpt-oss-120b |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 37.8% | 88.9% |
| LMArena Math | 1412 | 1388 |
| Omni-MATH | — | 68.8% |
| LiveBench Math | 69.3% | — |
| MATH Level 5 | 78.6% | — |
Knowledge gpt-oss-120b leads
GPT-4.5: 32.4 (#205), gpt-oss-120b: 42.3 (#89)
| Benchmark | GPT-4.5 | gpt-oss-120b |
|---|---|---|
| GPQA Diamond | 68.7% | 75.8% |
| Confabulations | 13.6% | 15.7% |
| LMArena Expert | 1394 | 1356 |
| Humanity's Last Exam | 5.4% | — |
| MMLU-Pro | — | 79.5% |
| Vectara Hallucination Rate | — | 14.2% |
| GPQA (HELM) | — | 68.4% |
Multimodal Not comparable
GPT-4.5: 37.0 (#68), gpt-oss-120b: —
| Benchmark | GPT-4.5 | gpt-oss-120b |
|---|---|---|
| LMArena Vision | 1195 | — |
| VPCT | 45% | — |
Multilingual GPT-4.5 leads
GPT-4.5: 52.7 (#78), gpt-oss-120b: 48.2 (#142)
| Benchmark | GPT-4.5 | gpt-oss-120b |
|---|---|---|
| LMArena Non-English | 1413 | 1350 |
| LMArena Chinese | 1421 | 1384 |
| LMArena French | 1417 | 1370 |
| LMArena German | 1457 | 1355 |
| LMArena Japanese | 1416 | 1332 |
| LMArena Korean | 1392 | 1283 |
| LMArena Russian | 1419 | 1343 |
| LMArena Spanish | — | 1388 |
Instruction Following GPT-4.5 leads
GPT-4.5: 72.8 (#129), gpt-oss-120b: 69.5 (#168)
| Benchmark | GPT-4.5 | gpt-oss-120b |
|---|---|---|
| LMArena Instruction Following | 1404 | 1318 |
| LiveBench Instruction Following | 72.3% | — |
| IFEval | — | 83.6% |
Long Context GPT-4.5 leads
GPT-4.5: 40.4 (#150), gpt-oss-120b: 31.4 (#273)
| Benchmark | GPT-4.5 | gpt-oss-120b |
|---|---|---|
| Fiction.LiveBench | 63.9% | 44.4% |
| LMArena Longer Query | 1406 | 1319 |
Writing & Preference GPT-4.5 leads
GPT-4.5: 57.1 (#129), gpt-oss-120b: 46.7 (#213)
| Benchmark | GPT-4.5 | gpt-oss-120b |
|---|---|---|
| LMArena Text | 1417 | 1366 |
| LMArena Creative Writing | 1394 | 1274 |
| Short-Story Creative Writing | 75.6% | 77.1% |
| EQ-Bench Creative Writing | 1258 | 961 |
| LMArena Multi-Turn | 1444 | 1340 |
| WildBench | — | 84.5% |
| LiveBench Language | 61.5% | — |
Frequently asked questions
Is GPT-4.5 better than gpt-oss-120b?
GPT-4.5 and gpt-oss-120b score almost the same on the Noometry Index (36.9 vs 36.0), so choose on price, context window or the category you care about most.
Is GPT-4.5 or gpt-oss-120b better for coding?
GPT-4.5 scores higher on coding benchmarks: 41.8 versus 33.2 in the Noometry coding category.
How many benchmarks do GPT-4.5 and gpt-oss-120b share?
26 benchmarks have published results for both models. GPT-4.5 has 42 scored results on Noometry and gpt-oss-120b has 48.