Model comparison
Claude Opus 5 vs gpt-oss-120b
Claude Opus 5 is the stronger model overall, scoring 67.8 to 36.3 on the Noometry Index. gpt-oss-120b costs 142× less per token, which makes it the better buy when Claude Opus 5's lead doesn't matter for your workload.
Last verified . 32 shared benchmarks.
Summary
- They share 32 benchmarks with published results for both. Claude Opus 5 scores higher in 9 categories and gpt-oss-120b in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Opus 5 leads 77.2 to 20.0.
- The biggest single-benchmark swing is APEX-Agents: 65.8% for Claude Opus 5 and 4.4% for gpt-oss-120b.
- gpt-oss-120b is cheaper at $0.037 / $0.17 per million input/output tokens, against $5 / $25 for Claude Opus 5.
- Claude Opus 5 accepts more context: 1M tokens versus 131K.
- gpt-oss-120b has downloadable open weights; the other is API-only.
Side by side
| Claude Opus 5 | gpt-oss-120b | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 67.8 | 36.3 |
| Released | 2026-07-24 | 2025-08-05 |
| Weights | Proprietary | Open |
| Context window | 1M | 131K |
| Max output | 128K | 41K |
| Input $ / M tokens | $5 | $0.037 |
| Output $ / M tokens | $25 | $0.17 |
| Results tracked | 57 | 48 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Opus 5 leads
Claude Opus 5: 67.5 (#5), gpt-oss-120b: 33.5 (#256)
| Benchmark | Claude Opus 5 | gpt-oss-120b |
|---|---|---|
| SciCode | 56.4% | 36% |
| WeirdML | 91.8% | 48.2% |
| LMArena Coding | 1534 | 1380 |
| ALE-Bench | 2,165 | 575.62 |
| DeepSWE | 73.6% | — |
| FrontierCode | 53.4% | — |
| SWE-bench Verified (bash only) | — | 26% |
| Aider Polyglot | — | 41.8% |
| CursorBench | 46.6% | — |
| LMArena WebDev | 1691 | — |
| FrontierSWE | 52% | — |
| AlgoTune | — | 1.41 |
Agentic & Tool Use Claude Opus 5 leads
Claude Opus 5: 55.6 (#1), gpt-oss-120b: 12.2 (#153)
| Benchmark | Claude Opus 5 | gpt-oss-120b |
|---|---|---|
| APEX-Agents | 65.8% | 4.4% |
| Vending-Bench 2 | 11,182 | -21.53 |
| Terminal-Bench | — | 18.7% |
| OSWorld 2.0 | 31.4% | — |
| τ²-bench Banking | 48.7% | — |
| PostTrainBench | 35% | — |
| BALROG | 63.4% | — |
| GBAEval | 79.6% | — |
| GDP.pdf | 24% | — |
| METR Time Horizons | — | 56.6% |
Reasoning Claude Opus 5 leads
Claude Opus 5: 77.2 (#4), gpt-oss-120b: 20.0 (#245)
| Benchmark | Claude Opus 5 | gpt-oss-120b |
|---|---|---|
| SimpleBench | 80.6% | 22.1% |
| CritPt | 29.1% | 1.1% |
| Chess Puzzles | 42% | 20% |
| LMArena Hard Prompts | 1526 | 1364 |
| Mystery Game Puzzles | 59% | 2% |
| DTBench | 97.9% | 76.3% |
| LMCA | 64.5% | 22.1% |
| Epoch Capabilities Index | 162.78 | 139.93 |
| ARC-AGI-2 | 90.4% | — |
| Kagi LLM Benchmark | — | 58.6% |
| NYT Connections (extended) | 94.3% | — |
| ARC-AGI-1 | 97.5% | — |
| EBR-Bench | 45.7% | — |
| Surface Evolver Bench | — | 25% |
| Bench to the Future 3 | 0.12 | — |
Math Claude Opus 5 leads
Claude Opus 5: 86.2 (#8), gpt-oss-120b: 52.5 (#50)
| Benchmark | Claude Opus 5 | gpt-oss-120b |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.9% | 88.9% |
| LMArena Math | 1531 | 1389 |
| FrontierMath (Tiers 1-3) | 85.6% | — |
| FrontierMath Tier 4 | 73.2% | — |
| ProofBench | 99% | — |
| Omni-MATH | — | 68.8% |
Knowledge Claude Opus 5 leads
Claude Opus 5: 66.8 (#9), gpt-oss-120b: 42.4 (#96)
| Benchmark | Claude Opus 5 | gpt-oss-120b |
|---|---|---|
| GPQA Diamond | 93.9% | 75.8% |
| LMArena Expert | 1557 | 1356 |
| SimpleQA Verified | 59.9% | — |
| MMLU-Pro | — | 79.5% |
| Confabulations | — | 15.7% |
| Vectara Hallucination Rate | — | 14.2% |
| GPQA (HELM) | — | 68.4% |
Multimodal Not comparable
Claude Opus 5: 50.8 (#8), gpt-oss-120b: —
| Benchmark | Claude Opus 5 | gpt-oss-120b |
|---|---|---|
| LMArena Vision | 1319 | — |
| Blueprint-Bench 2 | 30.4% | — |
| Furniture Assembly | 60.8% | — |
| LMArena Document | 1516 | — |
Multilingual Claude Opus 5 leads
Claude Opus 5: 58.8 (#4), gpt-oss-120b: 48.0 (#147)
| Benchmark | Claude Opus 5 | gpt-oss-120b |
|---|---|---|
| LMArena Non-English | 1501 | 1351 |
| LMArena Chinese | 1574 | 1385 |
| LMArena French | 1519 | 1369 |
| LMArena German | 1524 | 1353 |
| LMArena Japanese | 1516 | 1331 |
| LMArena Korean | 1521 | 1282 |
| LMArena Russian | 1507 | 1343 |
| LMArena Spanish | 1519 | 1389 |
Instruction Following Claude Opus 5 leads
Claude Opus 5: 79.2 (#7), gpt-oss-120b: 69.3 (#173)
| Benchmark | Claude Opus 5 | gpt-oss-120b |
|---|---|---|
| LMArena Instruction Following | 1517 | 1318 |
| IFEval | — | 83.6% |
Long Context Claude Opus 5 leads
Claude Opus 5: 46.5 (#21), gpt-oss-120b: 31.4 (#278)
| Benchmark | Claude Opus 5 | gpt-oss-120b |
|---|---|---|
| LMArena Longer Query | 1515 | 1319 |
| Fiction.LiveBench | — | 44.4% |
Writing & Preference Claude Opus 5 leads
Claude Opus 5: 79.2 (#1), gpt-oss-120b: 46.5 (#217)
| Benchmark | Claude Opus 5 | gpt-oss-120b |
|---|---|---|
| LMArena Text | 1507 | 1365 |
| LMArena Creative Writing | 1491 | 1275 |
| EQ-Bench Creative Writing | 2133 | 961 |
| LMArena Multi-Turn | 1499 | 1340 |
| Short-Story Creative Writing | — | 77.1% |
| WildBench | — | 84.5% |
| EQ-Bench 4 | 1385 | — |
Frequently asked questions
Is Claude Opus 5 better than gpt-oss-120b?
Claude Opus 5 is the stronger model overall, scoring 67.8 to 36.3 on the Noometry Index. gpt-oss-120b costs 142× less per token, which makes it the better buy when Claude Opus 5's lead doesn't matter for your workload.
Which is cheaper, Claude Opus 5 or gpt-oss-120b?
gpt-oss-120b is cheaper. It lists at $0.037 per million input tokens and $0.17 per million output tokens; Claude Opus 5 lists at $5 and $25.
Is Claude Opus 5 or gpt-oss-120b better for coding?
Claude Opus 5 scores higher on coding benchmarks: 67.5 versus 33.5 in the Noometry coding category.
Which has the bigger context window?
Claude Opus 5 does, with 1M tokens against 131K.
How many benchmarks do Claude Opus 5 and gpt-oss-120b share?
32 benchmarks have published results for both models. Claude Opus 5 has 57 scored results on Noometry and gpt-oss-120b has 48.