Model comparison
Gemma 2 9B vs gpt-oss-20b
gpt-oss-20b is the stronger model overall, scoring 32.1 to 25.7 on the Noometry Index.
Last verified . 20 shared benchmarks.
GoogleSummary
- They share 20 benchmarks with published results for both. Gemma 2 9B scores higher in 0 categories and gpt-oss-20b in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where gpt-oss-20b leads 37.6 to 9.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 0.6% for Gemma 2 9B and 65.3% for gpt-oss-20b.
Side by side
| Gemma 2 9B | gpt-oss-20b | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 25.7 | 32.1 |
| Released | 2024-06-24 | 2025-08-05 |
| Weights | Open | Open |
| Context window | — | 131K |
| Max output | — | 16K |
| Input $ / M tokens | — | $0.018 |
| Output $ / M tokens | — | $0.09 |
| Results tracked | 35 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding gpt-oss-20b leads
Gemma 2 9B: 29.0 (#299), gpt-oss-20b: 37.1 (#185)
| Benchmark | Gemma 2 9B | gpt-oss-20b |
|---|---|---|
| LMArena Coding | 1174 | 1307 |
| SciCode | — | 34.4% |
| WeirdML | — | 40.9% |
| BigCodeBench Instruct | 34.7% | — |
| LiveBench Coding | 22.5% | — |
| BigCodeBench Complete | 40.6% | — |
| ALE-Bench | — | 566.05 |
Agentic & Tool Use Not comparable
Gemma 2 9B: —, gpt-oss-20b: 9.2 (#146)
| Benchmark | Gemma 2 9B | gpt-oss-20b |
|---|---|---|
| Terminal-Bench | — | 3.4% |
Reasoning gpt-oss-20b leads
Gemma 2 9B: 15.3 (#306), gpt-oss-20b: 18.8 (#251)
| Benchmark | Gemma 2 9B | gpt-oss-20b |
|---|---|---|
| LMArena Hard Prompts | 1172 | 1274 |
| Epoch Capabilities Index | 119.78 | 137.81 |
| Kagi LLM Benchmark | — | 53.2% |
| CritPt | — | 1.4% |
| Chess Puzzles | — | 4% |
| LiveBench Reasoning | 15.2% | — |
| DTBench | — | 68% |
| LiveBench Data Analysis | 36.4% | — |
| LMCA | — | 14.5% |
| LiveBench | 28.7% | — |
| PIQA | 83.7% | — |
Math gpt-oss-20b leads
Gemma 2 9B: 9.3 (#312), gpt-oss-20b: 37.6 (#93)
| Benchmark | Gemma 2 9B | gpt-oss-20b |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0.6% | 65.3% |
| LMArena Math | 1183 | 1317 |
| Omni-MATH | — | 56.5% |
| LiveBench Math | 19.8% | — |
| MATH Level 5 | 21% | — |
| GSM8K | 84.9% | — |
Knowledge gpt-oss-20b leads
Gemma 2 9B: 9.7 (#300), gpt-oss-20b: 34.6 (#186)
| Benchmark | Gemma 2 9B | gpt-oss-20b |
|---|---|---|
| GPQA Diamond | 27.5% | 60.8% |
| LMArena Expert | 1147 | 1258 |
| MMLU-Pro | — | 74% |
| GPQA (HELM) | — | 59.4% |
| BoolQ | 85.7% | — |
| MMLU | 72.1% | — |
Multilingual gpt-oss-20b leads
Gemma 2 9B: 36.8 (#233), gpt-oss-20b: 42.4 (#192)
| Benchmark | Gemma 2 9B | gpt-oss-20b |
|---|---|---|
| LMArena Non-English | 1188 | 1269 |
| LMArena Chinese | 1185 | 1311 |
| LMArena German | 1186 | 1249 |
| LMArena Japanese | 1144 | 1245 |
| LMArena Korean | 1137 | 1239 |
| LMArena Russian | 1200 | 1279 |
| LMArena Spanish | 1200 | 1267 |
| LMArena French | 1189 | — |
Instruction Following gpt-oss-20b leads
Gemma 2 9B: 57.8 (#264), gpt-oss-20b: 62.0 (#235)
| Benchmark | Gemma 2 9B | gpt-oss-20b |
|---|---|---|
| LMArena Instruction Following | 1178 | 1236 |
| LiveBench Instruction Following | 52.6% | — |
| IFEval | — | 73.2% |
Long Context gpt-oss-20b leads
Gemma 2 9B: 36.3 (#228), gpt-oss-20b: 37.9 (#203)
| Benchmark | Gemma 2 9B | gpt-oss-20b |
|---|---|---|
| LMArena Longer Query | 1197 | 1251 |
Writing & Preference gpt-oss-20b leads
Gemma 2 9B: 32.3 (#276), gpt-oss-20b: 35.8 (#260)
| Benchmark | Gemma 2 9B | gpt-oss-20b |
|---|---|---|
| LMArena Text | 1207 | 1288 |
| LMArena Creative Writing | 1206 | 1201 |
| EQ-Bench Creative Writing | 841 | 666 |
| LMArena Multi-Turn | 1193 | 1268 |
| WildBench | — | 73.7% |
| LiveBench Language | 25.5% | — |
Frequently asked questions
Is Gemma 2 9B better than gpt-oss-20b?
gpt-oss-20b is the stronger model overall, scoring 32.1 to 25.7 on the Noometry Index.
Is Gemma 2 9B or gpt-oss-20b better for coding?
gpt-oss-20b scores higher on coding benchmarks: 37.1 versus 29.0 in the Noometry coding category.
How many benchmarks do Gemma 2 9B and gpt-oss-20b share?
20 benchmarks have published results for both models. Gemma 2 9B has 35 scored results on Noometry and gpt-oss-20b has 34.