Model comparison
Gemma 2 9B vs GPT-4o mini
Gemma 2 9B and GPT-4o mini score almost the same on the Noometry Index (25.9 vs 25.5), so choose on price, context window or the category you care about most.
Last verified . 35 shared benchmarks.
GoogleSummary
- They share 35 benchmarks with published results for both. Gemma 2 9B scores higher in 2 categories and GPT-4o mini in 6 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where GPT-4o mini leads 17.7 to 9.7.
- The biggest single-benchmark swing is MATH Level 5: 21% for Gemma 2 9B and 52.6% for GPT-4o mini.
- Gemma 2 9B has downloadable open weights; the other is API-only.
Side by side
| Gemma 2 9B | GPT-4o mini | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 25.9 | 25.5 |
| Released | 2024-06-24 | 2024-07-18 |
| Weights | Open | Proprietary |
| Context window | — | 128K |
| Max output | — | 16K |
| Input $ / M tokens | — | $0.15 |
| Output $ / M tokens | — | $0.60 |
| Results tracked | 35 | 60 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 2 9B leads
Gemma 2 9B: 29.4 (#304), GPT-4o mini: 22.0 (#335)
| Benchmark | Gemma 2 9B | GPT-4o mini |
|---|---|---|
| BigCodeBench Instruct | 34.7% | 46.1% |
| LiveBench Coding | 22.5% | 43.1% |
| LMArena Coding | 1173 | 1290 |
| BigCodeBench Complete | 40.6% | 57.4% |
| Aider Polyglot | — | 3.6% |
| WeirdML | — | 11.8% |
| HumanEval+ | — | 83.5% |
| MBPP+ | — | 72.2% |
Agentic & Tool Use Not comparable
Gemma 2 9B: —, GPT-4o mini: 27.5 (#101)
| Benchmark | Gemma 2 9B | GPT-4o mini |
|---|---|---|
| BALROG | — | 17.4% |
Reasoning Gemma 2 9B leads
Gemma 2 9B: 15.9 (#309), GPT-4o mini: 8.7 (#347)
| Benchmark | Gemma 2 9B | GPT-4o mini |
|---|---|---|
| LiveBench Reasoning | 15.2% | 32.8% |
| LMArena Hard Prompts | 1171 | 1267 |
| LiveBench Data Analysis | 36.4% | 50% |
| Epoch Capabilities Index | 119.83 | 126.56 |
| LiveBench | 28.7% | 41.3% |
| PIQA | 83.7% | 88.7% |
| ARC-AGI-2 | — | 0% |
| SimpleBench | — | 10.7% |
| Kagi LLM Benchmark | — | 28.8% |
| Chess Puzzles | — | 0% |
| Mystery Game Puzzles | — | 12% |
| DTBench | — | 54.4% |
| LMCA | — | 10.4% |
Math Too close to call
Gemma 2 9B: 9.9 (#318), GPT-4o mini: 10.4 (#314)
| Benchmark | Gemma 2 9B | GPT-4o mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0.6% | 6.9% |
| LiveBench Math | 19.8% | 36.3% |
| LMArena Math | 1183 | 1267 |
| MATH Level 5 | 21% | 52.6% |
| GSM8K | 84.9% | 91.3% |
| FrontierMath (Tiers 1-3) | — | 0.7% |
| Omni-MATH | — | 28% |
Knowledge GPT-4o mini leads
Gemma 2 9B: 9.7 (#305), GPT-4o mini: 17.7 (#284)
| Benchmark | Gemma 2 9B | GPT-4o mini |
|---|---|---|
| GPQA Diamond | 27.5% | 37.7% |
| LMArena Expert | 1147 | 1235 |
| BoolQ | 85.7% | 88.7% |
| MMLU | 72.1% | 81.8% |
| SimpleQA Verified | — | 8.3% |
| MMLU-Pro | — | 60.3% |
| Confabulations | — | 37.2% |
| GPQA (HELM) | — | 36.8% |
Multimodal Not comparable
Gemma 2 9B: —, GPT-4o mini: 25.9 (#122)
| Benchmark | Gemma 2 9B | GPT-4o mini |
|---|---|---|
| LMArena Vision | — | 1066 |
| Video-MME | — | 64.8% |
| GeoBench | — | 64% |
| VPCT | — | 34% |
Multilingual GPT-4o mini leads
Gemma 2 9B: 36.6 (#238), GPT-4o mini: 42.0 (#199)
| Benchmark | Gemma 2 9B | GPT-4o mini |
|---|---|---|
| LMArena Non-English | 1188 | 1266 |
| LMArena Chinese | 1185 | 1265 |
| LMArena French | 1190 | 1297 |
| LMArena German | 1186 | 1272 |
| LMArena Japanese | 1144 | 1216 |
| LMArena Korean | 1137 | 1195 |
| LMArena Russian | 1200 | 1275 |
| LMArena Spanish | 1200 | 1276 |
Instruction Following GPT-4o mini leads
Gemma 2 9B: 57.6 (#269), GPT-4o mini: 61.9 (#239)
| Benchmark | Gemma 2 9B | GPT-4o mini |
|---|---|---|
| LiveBench Instruction Following | 52.6% | 56.8% |
| LMArena Instruction Following | 1178 | 1258 |
| IFEval | — | 78.2% |
Long Context GPT-4o mini leads
Gemma 2 9B: 36.3 (#233), GPT-4o mini: 39.1 (#186)
| Benchmark | Gemma 2 9B | GPT-4o mini |
|---|---|---|
| LMArena Longer Query | 1197 | 1289 |
Writing & Preference GPT-4o mini leads
Gemma 2 9B: 32.1 (#281), GPT-4o mini: 39.5 (#248)
| Benchmark | Gemma 2 9B | GPT-4o mini |
|---|---|---|
| LMArena Text | 1207 | 1286 |
| LMArena Creative Writing | 1206 | 1268 |
| EQ-Bench Creative Writing | 841 | 873 |
| LMArena Multi-Turn | 1193 | 1285 |
| LiveBench Language | 25.5% | 28.6% |
| Short-Story Creative Writing | — | 67.2% |
| WildBench | — | 79.1% |
Frequently asked questions
Is Gemma 2 9B better than GPT-4o mini?
Gemma 2 9B and GPT-4o mini score almost the same on the Noometry Index (25.9 vs 25.5), so choose on price, context window or the category you care about most.
Is Gemma 2 9B or GPT-4o mini better for coding?
Gemma 2 9B scores higher on coding benchmarks: 29.4 versus 22.0 in the Noometry coding category.
How many benchmarks do Gemma 2 9B and GPT-4o mini share?
35 benchmarks have published results for both models. Gemma 2 9B has 35 scored results on Noometry and GPT-4o mini has 60.