Model comparison
Gemma 4 31B IT vs gpt-oss-120b
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 36.3 on the Noometry Index. gpt-oss-120b costs 2.2× less per token, which makes it the better buy when Gemma 4 31B IT's lead doesn't matter for your workload.
Last verified . 28 shared benchmarks.
GoogleSummary
- They share 28 benchmarks with published results for both. Gemma 4 31B IT scores higher in 6 categories and gpt-oss-120b in 2 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Gemma 4 31B IT leads 60.5 to 46.5.
- The biggest single-benchmark swing is LMCA: 39.3% for Gemma 4 31B IT and 22.1% for gpt-oss-120b.
- gpt-oss-120b is cheaper at $0.037 / $0.17 per million input/output tokens, against $0.09 / $0.34 for Gemma 4 31B IT.
- Gemma 4 31B IT accepts more context: 262K tokens versus 131K.
Side by side
| Gemma 4 31B IT | gpt-oss-120b | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 43.5 | 36.3 |
| Released | 2026-04-02 | 2025-08-05 |
| Weights | Open | Open |
| Context window | 262K | 131K |
| Max output | 33K | 41K |
| Input $ / M tokens | $0.09 | $0.037 |
| Output $ / M tokens | $0.34 | $0.17 |
| Results tracked | 35 | 48 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 4 31B IT leads
Gemma 4 31B IT: 42.3 (#108), gpt-oss-120b: 33.5 (#256)
| Benchmark | Gemma 4 31B IT | gpt-oss-120b |
|---|---|---|
| SciCode | 43.4% | 36% |
| WeirdML | 52.3% | 48.2% |
| LMArena Coding | 1459 | 1380 |
| ALE-Bench | 925.5 | 575.62 |
| SWE-bench Verified (bash only) | — | 26% |
| Aider Polyglot | — | 41.8% |
| LMArena WebDev | 1366 | — |
| AlgoTune | — | 1.41 |
Agentic & Tool Use Not comparable
Gemma 4 31B IT: —, gpt-oss-120b: 12.2 (#153)
| Benchmark | Gemma 4 31B IT | gpt-oss-120b |
|---|---|---|
| Terminal-Bench | — | 18.7% |
| APEX-Agents | — | 4.4% |
| METR Time Horizons | — | 56.6% |
| Vending-Bench 2 | — | -21.53 |
Reasoning Gemma 4 31B IT leads
Gemma 4 31B IT: 27.2 (#122), gpt-oss-120b: 20.0 (#245)
| Benchmark | Gemma 4 31B IT | gpt-oss-120b |
|---|---|---|
| Kagi LLM Benchmark | 63.5% | 58.6% |
| CritPt | 1.4% | 1.1% |
| Chess Puzzles | 5% | 20% |
| LMArena Hard Prompts | 1448 | 1364 |
| DTBench | 82.7% | 76.3% |
| LMCA | 39.3% | 22.1% |
| Surface Evolver Bench | 30.6% | 25% |
| Epoch Capabilities Index | 142.74 | 139.93 |
| SimpleBench | — | 22.1% |
| NYT Connections (extended) | 70.6% | — |
| Thematic Generalization | 53% | — |
| Mystery Game Puzzles | — | 2% |
Math gpt-oss-120b leads
Gemma 4 31B IT: 43.2 (#81), gpt-oss-120b: 52.5 (#50)
| Benchmark | Gemma 4 31B IT | gpt-oss-120b |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.3% | 88.9% |
| LMArena Math | 1465 | 1389 |
| Omni-MATH | — | 68.8% |
Knowledge gpt-oss-120b leads
Gemma 4 31B IT: 37.9 (#151), gpt-oss-120b: 42.4 (#96)
| Benchmark | Gemma 4 31B IT | gpt-oss-120b |
|---|---|---|
| GPQA Diamond | 75.8% | 75.8% |
| Vectara Hallucination Rate | 7.4% | 14.2% |
| LMArena Expert | 1465 | 1356 |
| SimpleQA Verified | 10.4% | — |
| MMLU-Pro | — | 79.5% |
| Confabulations | — | 15.7% |
| GPQA (HELM) | — | 68.4% |
Multimodal Not comparable
Gemma 4 31B IT: 41.6 (#34), gpt-oss-120b: —
| Benchmark | Gemma 4 31B IT | gpt-oss-120b |
|---|---|---|
| LMArena Vision | 1277 | — |
| LMArena Document | 1425 | — |
Multilingual Gemma 4 31B IT leads
Gemma 4 31B IT: 53.8 (#57), gpt-oss-120b: 48.0 (#147)
| Benchmark | Gemma 4 31B IT | gpt-oss-120b |
|---|---|---|
| LMArena Non-English | 1431 | 1351 |
| LMArena Chinese | 1476 | 1385 |
| LMArena French | 1435 | 1369 |
| LMArena Russian | 1460 | 1343 |
| LMArena Spanish | 1444 | 1389 |
| LMArena German | — | 1353 |
| LMArena Japanese | — | 1331 |
| LMArena Korean | — | 1282 |
Instruction Following Gemma 4 31B IT leads
Gemma 4 31B IT: 75.5 (#61), gpt-oss-120b: 69.3 (#173)
| Benchmark | Gemma 4 31B IT | gpt-oss-120b |
|---|---|---|
| LMArena Instruction Following | 1433 | 1318 |
| IFEval | — | 83.6% |
Long Context Gemma 4 31B IT leads
Gemma 4 31B IT: 44.2 (#71), gpt-oss-120b: 31.4 (#278)
| Benchmark | Gemma 4 31B IT | gpt-oss-120b |
|---|---|---|
| LMArena Longer Query | 1446 | 1319 |
| Fiction.LiveBench | — | 44.4% |
Writing & Preference Gemma 4 31B IT leads
Gemma 4 31B IT: 60.5 (#96), gpt-oss-120b: 46.5 (#217)
| Benchmark | Gemma 4 31B IT | gpt-oss-120b |
|---|---|---|
| LMArena Text | 1443 | 1365 |
| LMArena Creative Writing | 1415 | 1275 |
| EQ-Bench Creative Writing | 1368 | 961 |
| LMArena Multi-Turn | 1452 | 1340 |
| Short-Story Creative Writing | — | 77.1% |
| WildBench | — | 84.5% |
| EQ-Bench 4 | 1120 | — |
Frequently asked questions
Is Gemma 4 31B IT better than gpt-oss-120b?
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 36.3 on the Noometry Index. gpt-oss-120b costs 2.2× less per token, which makes it the better buy when Gemma 4 31B IT's lead doesn't matter for your workload.
Which is cheaper, Gemma 4 31B IT 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; Gemma 4 31B IT lists at $0.09 and $0.34.
Is Gemma 4 31B IT or gpt-oss-120b better for coding?
Gemma 4 31B IT scores higher on coding benchmarks: 42.3 versus 33.5 in the Noometry coding category.
Which has the bigger context window?
Gemma 4 31B IT does, with 262K tokens against 131K.
How many benchmarks do Gemma 4 31B IT and gpt-oss-120b share?
28 benchmarks have published results for both models. Gemma 4 31B IT has 35 scored results on Noometry and gpt-oss-120b has 48.