Model comparison
Gemma 2 9B vs GPT-5 Mini
GPT-5 Mini is the stronger model overall, scoring 41.8 to 25.9 on the Noometry Index.
Last verified . 22 shared benchmarks.
GoogleSummary
- They share 22 benchmarks with published results for both. Gemma 2 9B scores higher in 0 categories and GPT-5 Mini in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5 Mini leads 46.7 to 9.9.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 0.6% for Gemma 2 9B and 86.7% for GPT-5 Mini.
- Gemma 2 9B has downloadable open weights; the other is API-only.
Side by side
| Gemma 2 9B | GPT-5 Mini | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 25.9 | 41.8 |
| Released | 2024-06-24 | 2025-08-07 |
| Weights | Open | Proprietary |
| Context window | — | 400K |
| Max output | — | 128K |
| Input $ / M tokens | — | $0.25 |
| Output $ / M tokens | — | $2 |
| Results tracked | 35 | 60 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5 Mini leads
Gemma 2 9B: 29.4 (#304), GPT-5 Mini: 40.1 (#146)
| Benchmark | Gemma 2 9B | GPT-5 Mini |
|---|---|---|
| LMArena Coding | 1173 | 1406 |
| SWE-bench Verified | — | 64.7% |
| SWE-bench Verified (bash only) | — | 59.8% |
| SWE-bench Multilingual | — | 39.7% |
| SciCode | — | 39.2% |
| WeirdML | — | 52.7% |
| BigCodeBench Instruct | 34.7% | — |
| LiveBench Coding | 22.5% | — |
| BigCodeBench Complete | 40.6% | — |
| ALE-Bench | — | 799.77 |
| AlgoTune | — | 1.38 |
Agentic & Tool Use Not comparable
Gemma 2 9B: —, GPT-5 Mini: 31.1 (#70)
| Benchmark | Gemma 2 9B | GPT-5 Mini |
|---|---|---|
| Terminal-Bench | — | 34.8% |
| Berkeley Function Calling Leaderboard | — | 55.5% |
| Vending-Bench 2 | — | -31.18 |
Reasoning GPT-5 Mini leads
Gemma 2 9B: 15.9 (#309), GPT-5 Mini: 23.9 (#168)
| Benchmark | Gemma 2 9B | GPT-5 Mini |
|---|---|---|
| LMArena Hard Prompts | 1171 | 1380 |
| Epoch Capabilities Index | 119.83 | 145.52 |
| ARC-AGI-2 | — | 4.4% |
| Kagi LLM Benchmark | — | 70.3% |
| ARC-AGI-1 | — | 54.3% |
| CritPt | — | 0% |
| Chess Puzzles | — | 30% |
| EnigmaEval | — | 8.2% |
| LiveBench Reasoning | 15.2% | — |
| Mystery Game Puzzles | — | 10% |
| DTBench | — | 80.5% |
| LiveBench Data Analysis | 36.4% | — |
| LMCA | — | 34.2% |
| ForecastBench | — | 61 |
| LiveBench | 28.7% | — |
| PIQA | 83.7% | — |
Math GPT-5 Mini leads
Gemma 2 9B: 9.9 (#318), GPT-5 Mini: 46.7 (#69)
| Benchmark | Gemma 2 9B | GPT-5 Mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0.6% | 86.7% |
| LMArena Math | 1183 | 1378 |
| MATH Level 5 | 21% | 97.8% |
| FrontierMath (Tiers 1-3) | — | 46.7% |
| FrontierMath Tier 4 | — | 12.2% |
| ProofBench | — | 9% |
| Omni-MATH | — | 72.2% |
| LiveBench Math | 19.8% | — |
| FrontierMath (Feb 2025 set) | — | 27.2% |
| FrontierMath Tier 4 (v1) | — | 6.3% |
| GSM8K | 84.9% | — |
Knowledge GPT-5 Mini leads
Gemma 2 9B: 9.7 (#305), GPT-5 Mini: 45.6 (#86)
| Benchmark | Gemma 2 9B | GPT-5 Mini |
|---|---|---|
| GPQA Diamond | 27.5% | 75% |
| LMArena Expert | 1147 | 1379 |
| Humanity's Last Exam | — | 19.4% |
| SimpleQA Verified | — | 21.6% |
| MMLU-Pro | — | 83.5% |
| Confabulations | — | 13.3% |
| Vectara Hallucination Rate | — | 12.9% |
| GPQA (HELM) | — | 75.6% |
| BoolQ | 85.7% | — |
| MMLU | 72.1% | — |
Multimodal Not comparable
Gemma 2 9B: —, GPT-5 Mini: 35.6 (#85)
| Benchmark | Gemma 2 9B | GPT-5 Mini |
|---|---|---|
| LMArena Vision | — | 1202 |
| VPCT | — | 40.2% |
Multilingual GPT-5 Mini leads
Gemma 2 9B: 36.6 (#238), GPT-5 Mini: 48.9 (#137)
| Benchmark | Gemma 2 9B | GPT-5 Mini |
|---|---|---|
| LMArena Non-English | 1188 | 1363 |
| LMArena Chinese | 1185 | 1385 |
| LMArena French | 1190 | 1386 |
| LMArena German | 1186 | 1366 |
| LMArena Japanese | 1144 | 1341 |
| LMArena Korean | 1137 | 1308 |
| LMArena Russian | 1200 | 1362 |
| LMArena Spanish | 1200 | 1355 |
Instruction Following GPT-5 Mini leads
Gemma 2 9B: 57.6 (#269), GPT-5 Mini: 76.2 (#46)
| Benchmark | Gemma 2 9B | GPT-5 Mini |
|---|---|---|
| LMArena Instruction Following | 1178 | 1357 |
| LiveBench Instruction Following | 52.6% | — |
| IFEval | — | 92.7% |
Long Context GPT-5 Mini leads
Gemma 2 9B: 36.3 (#233), GPT-5 Mini: 41.9 (#132)
| Benchmark | Gemma 2 9B | GPT-5 Mini |
|---|---|---|
| LMArena Longer Query | 1197 | 1355 |
| Fiction.LiveBench | — | 69.4% |
Writing & Preference GPT-5 Mini leads
Gemma 2 9B: 32.1 (#281), GPT-5 Mini: 55.2 (#148)
| Benchmark | Gemma 2 9B | GPT-5 Mini |
|---|---|---|
| LMArena Text | 1207 | 1373 |
| LMArena Creative Writing | 1206 | 1325 |
| EQ-Bench Creative Writing | 841 | 1313 |
| LMArena Multi-Turn | 1193 | 1363 |
| Short-Story Creative Writing | — | 83.1% |
| WildBench | — | 85.5% |
| LiveBench Language | 25.5% | — |
Frequently asked questions
Is Gemma 2 9B better than GPT-5 Mini?
GPT-5 Mini is the stronger model overall, scoring 41.8 to 25.9 on the Noometry Index.
Is Gemma 2 9B or GPT-5 Mini better for coding?
GPT-5 Mini scores higher on coding benchmarks: 40.1 versus 29.4 in the Noometry coding category.
How many benchmarks do Gemma 2 9B and GPT-5 Mini share?
22 benchmarks have published results for both models. Gemma 2 9B has 35 scored results on Noometry and GPT-5 Mini has 60.