Model comparison
Gemma 2B vs GPT-4 Turbo
GPT-4 Turbo is the stronger model overall, scoring 30.3 to 29.1 on the Noometry Index.
Last verified . 15 shared benchmarks.
GoogleSummary
- They share 15 benchmarks with published results for both. Gemma 2B scores higher in 2 categories and GPT-4 Turbo in 5 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GPT-4 Turbo leads 47.9 to 24.1.
- Gemma 2B has downloadable open weights; the other is API-only.
Side by side
| Gemma 2B | GPT-4 Turbo | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 29.1 | 30.3 |
| 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 | 23 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-4 Turbo leads
Gemma 2B: 28.9 (#300), GPT-4 Turbo: 33.4 (#247)
| Benchmark | Gemma 2B | GPT-4 Turbo |
|---|---|---|
| LMArena Coding | 1010 | 1268 |
| HumanEval+ | 20.7% | 86.6% |
| MBPP+ | 34.1% | 73.3% |
| WeirdML | — | 18% |
| BigCodeBench Instruct | — | 48.2% |
| BigCodeBench Complete | — | 58.2% |
Agentic & Tool Use Not comparable
Gemma 2B: —, GPT-4 Turbo: —
| Benchmark | Gemma 2B | GPT-4 Turbo |
|---|---|---|
| METR Time Horizons | — | 36.7% |
Reasoning Gemma 2B leads
Gemma 2B: 17.9 (#275), GPT-4 Turbo: 14.9 (#311)
| Benchmark | Gemma 2B | GPT-4 Turbo |
|---|---|---|
| LMArena Hard Prompts | 989 | 1251 |
| Epoch Capabilities Index | 93.57 | 127.25 |
| SimpleBench | — | 25.1% |
| Chess Puzzles | — | 6% |
| DTBench | — | 61.6% |
| LMCA | — | 9.8% |
| BIG-Bench Hard | 35.2% | — |
| ForecastBench | — | 59.4 |
| HellaSwag | 71.4% | — |
| PIQA | 77.3% | — |
| WinoGrande | 65.4% | — |
Math Gemma 2B leads
Gemma 2B: 27.6 (#237), GPT-4 Turbo: 8.6 (#316)
| Benchmark | Gemma 2B | GPT-4 Turbo |
|---|---|---|
| LMArena Math | 1009 | 1272 |
| FrontierMath (Tiers 1-3) | — | 0.7% |
| OTIS Mock AIME 2024-2025 | — | 6.7% |
| MATH Level 5 | — | 46.7% |
| GSM8K | 17.7% | — |
Knowledge Not comparable
Gemma 2B: —, GPT-4 Turbo: 24.3 (#263)
| Benchmark | Gemma 2B | GPT-4 Turbo |
|---|---|---|
| MMLU | 42.3% | 81.3% |
| GPQA Diamond | — | 46.6% |
| Confabulations | — | 28.4% |
| LMArena Expert | — | 1223 |
| ARC (AI2) Challenge | 42.1% | — |
| BoolQ | 69.4% | — |
| TriviaQA | 53.2% | — |
Multimodal Not comparable
Gemma 2B: —, GPT-4 Turbo: 30.0 (#104)
| Benchmark | Gemma 2B | GPT-4 Turbo |
|---|---|---|
| LMArena Vision | — | 1090 |
Multilingual GPT-4 Turbo leads
Gemma 2B: 23.1 (#289), GPT-4 Turbo: 40.7 (#211)
| Benchmark | Gemma 2B | GPT-4 Turbo |
|---|---|---|
| LMArena Non-English | 958 | 1245 |
| LMArena Chinese | 986 | 1242 |
| LMArena Russian | 937 | 1259 |
| LMArena French | — | 1275 |
| LMArena German | — | 1259 |
| LMArena Japanese | — | 1194 |
| LMArena Korean | — | 1187 |
| LMArena Spanish | — | 1260 |
Instruction Following GPT-4 Turbo leads
Gemma 2B: 48.8 (#297), GPT-4 Turbo: 66.0 (#211)
| Benchmark | Gemma 2B | GPT-4 Turbo |
|---|---|---|
| LMArena Instruction Following | 971 | 1250 |
Long Context GPT-4 Turbo leads
Gemma 2B: 29.9 (#286), GPT-4 Turbo: 38.0 (#200)
| Benchmark | Gemma 2B | GPT-4 Turbo |
|---|---|---|
| LMArena Longer Query | 981 | 1254 |
Writing & Preference GPT-4 Turbo leads
Gemma 2B: 24.1 (#303), GPT-4 Turbo: 47.9 (#201)
| Benchmark | Gemma 2B | GPT-4 Turbo |
|---|---|---|
| LMArena Text | 1002 | 1272 |
| LMArena Creative Writing | 986 | 1269 |
| LMArena Multi-Turn | 945 | 1267 |
Frequently asked questions
Is Gemma 2B better than GPT-4 Turbo?
GPT-4 Turbo is the stronger model overall, scoring 30.3 to 29.1 on the Noometry Index.
Is Gemma 2B or GPT-4 Turbo better for coding?
GPT-4 Turbo scores higher on coding benchmarks: 33.4 versus 28.9 in the Noometry coding category.
How many benchmarks do Gemma 2B and GPT-4 Turbo share?
15 benchmarks have published results for both models. Gemma 2B has 23 scored results on Noometry and GPT-4 Turbo has 36.