Model comparison
Gemma 7B vs GPT-4 Turbo
Gemma 7B and GPT-4 Turbo score almost the same on the Noometry Index (30.0 vs 30.5), so choose on price, context window or the category you care about most.
Last verified . 17 shared benchmarks.
GoogleSummary
- They share 17 benchmarks with published results for both. Gemma 7B scores higher in 3 categories and GPT-4 Turbo in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Gemma 7B leads 31.2 to 9.0.
- Gemma 7B has downloadable open weights; the other is API-only.
Side by side
| Gemma 7B | GPT-4 Turbo | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 30.0 | 30.5 |
| Released | 2024-02-21 | 2023-11-06 |
| Weights | Open | Proprietary |
| Context window | — | 128K |
| Max output | — | 4K |
| Input $ / M tokens | — | $10 |
| Output $ / M tokens | — | $30 |
| Results tracked | 27 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-4 Turbo leads
Gemma 7B: 30.5 (#294), GPT-4 Turbo: 33.8 (#249)
| Benchmark | Gemma 7B | GPT-4 Turbo |
|---|---|---|
| LMArena Coding | 1048 | 1268 |
| HumanEval+ | 28.7% | 86.6% |
| MBPP+ | 43.4% | 73.3% |
| WeirdML | — | 18% |
| BigCodeBench Instruct | — | 48.2% |
| BigCodeBench Complete | — | 58.2% |
Agentic & Tool Use Not comparable
Gemma 7B: —, GPT-4 Turbo: —
| Benchmark | Gemma 7B | GPT-4 Turbo |
|---|---|---|
| METR Time Horizons | — | 36.7% |
Reasoning Gemma 7B leads
Gemma 7B: 19.9 (#249), GPT-4 Turbo: 15.3 (#317)
| Benchmark | Gemma 7B | GPT-4 Turbo |
|---|---|---|
| LMArena Hard Prompts | 1042 | 1251 |
| Epoch Capabilities Index | 111.99 | 127.25 |
| SimpleBench | — | 25.1% |
| Chess Puzzles | — | 6% |
| DTBench | — | 61.6% |
| LMCA | — | 9.8% |
| Adversarial NLI | 48.7% | — |
| BIG-Bench Hard | 55.1% | — |
| ForecastBench | — | 59.4 |
| HellaSwag | 82.2% | — |
| PIQA | 81.2% | — |
| WinoGrande | 79% | — |
Math Gemma 7B leads
Gemma 7B: 31.2 (#228), GPT-4 Turbo: 9.0 (#322)
| Benchmark | Gemma 7B | GPT-4 Turbo |
|---|---|---|
| LMArena Math | 1066 | 1272 |
| FrontierMath (Tiers 1-3) | — | 0.7% |
| OTIS Mock AIME 2024-2025 | — | 6.7% |
| MATH Level 5 | — | 46.7% |
| GSM8K | 46.4% | — |
Knowledge Gemma 7B leads
Gemma 7B: 27.3 (#252), GPT-4 Turbo: 24.3 (#268)
| Benchmark | Gemma 7B | GPT-4 Turbo |
|---|---|---|
| LMArena Expert | 1001 | 1223 |
| MMLU | 66.1% | 81.3% |
| GPQA Diamond | — | 46.6% |
| Confabulations | — | 28.4% |
| ARC (AI2) Challenge | 78.3% | — |
| BoolQ | 83.2% | — |
| OpenBookQA | 78.6% | — |
| TriviaQA | 72.3% | — |
Multimodal Not comparable
Gemma 7B: —, GPT-4 Turbo: 30.6 (#110)
| Benchmark | Gemma 7B | GPT-4 Turbo |
|---|---|---|
| LMArena Vision | — | 1090 |
Multilingual GPT-4 Turbo leads
Gemma 7B: 25.1 (#287), GPT-4 Turbo: 40.5 (#216)
| Benchmark | Gemma 7B | GPT-4 Turbo |
|---|---|---|
| LMArena Non-English | 999 | 1245 |
| LMArena Chinese | 1035 | 1242 |
| LMArena French | 1025 | 1276 |
| LMArena Russian | 993 | 1259 |
| LMArena German | — | 1259 |
| LMArena Japanese | — | 1194 |
| LMArena Korean | — | 1187 |
| LMArena Spanish | — | 1260 |
Instruction Following GPT-4 Turbo leads
Gemma 7B: 51.5 (#295), GPT-4 Turbo: 65.8 (#216)
| Benchmark | Gemma 7B | GPT-4 Turbo |
|---|---|---|
| LMArena Instruction Following | 1017 | 1249 |
Long Context GPT-4 Turbo leads
Gemma 7B: 31.1 (#282), GPT-4 Turbo: 38.0 (#206)
| Benchmark | Gemma 7B | GPT-4 Turbo |
|---|---|---|
| LMArena Longer Query | 1022 | 1254 |
Writing & Preference GPT-4 Turbo leads
Gemma 7B: 27.1 (#302), GPT-4 Turbo: 47.7 (#206)
| Benchmark | Gemma 7B | GPT-4 Turbo |
|---|---|---|
| LMArena Text | 1056 | 1272 |
| LMArena Creative Writing | 1024 | 1269 |
| LMArena Multi-Turn | 963 | 1267 |
Frequently asked questions
Is Gemma 7B better than GPT-4 Turbo?
Gemma 7B and GPT-4 Turbo score almost the same on the Noometry Index (30.0 vs 30.5), so choose on price, context window or the category you care about most.
Is Gemma 7B or GPT-4 Turbo better for coding?
GPT-4 Turbo scores higher on coding benchmarks: 33.8 versus 30.5 in the Noometry coding category.
How many benchmarks do Gemma 7B and GPT-4 Turbo share?
17 benchmarks have published results for both models. Gemma 7B has 27 scored results on Noometry and GPT-4 Turbo has 36.