Model comparison
Gemma 7B vs Mixtral 8x7B
Gemma 7B is the stronger model overall, scoring 30.0 to 27.1 on the Noometry Index.
Last verified . 25 shared benchmarks.
Google
Mistral AISummary
- They share 25 benchmarks with published results for both. Gemma 7B scores higher in 4 categories and Mixtral 8x7B in 4 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Gemma 7B leads 27.3 to 11.0.
Side by side
| Gemma 7B | Mixtral 8x7B | |
|---|---|---|
| Provider | Mistral AI | |
| Noometry Index | 30.0 | 27.1 |
| Released | 2024-02-21 | 2023-12-11 |
| Weights | Open | Open |
| Context window | — | 32K |
| Max output | — | 32K |
| Input $ / M tokens | — | $0.70 |
| Output $ / M tokens | — | $0.70 |
| Results tracked | 27 | 38 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mixtral 8x7B leads
Gemma 7B: 30.5 (#294), Mixtral 8x7B: 32.8 (#269)
| Benchmark | Gemma 7B | Mixtral 8x7B |
|---|---|---|
| LMArena Coding | 1048 | 1126 |
| HumanEval+ | 28.7% | 39.6% |
| MBPP+ | 43.4% | 49.7% |
Reasoning Gemma 7B leads
Gemma 7B: 19.9 (#249), Mixtral 8x7B: 18.2 (#285)
| Benchmark | Gemma 7B | Mixtral 8x7B |
|---|---|---|
| LMArena Hard Prompts | 1042 | 1115 |
| Adversarial NLI | 48.7% | 55.2% |
| Epoch Capabilities Index | 111.99 | 118.47 |
| HellaSwag | 82.2% | 86.7% |
| PIQA | 81.2% | 83.6% |
| WinoGrande | 79% | 77.2% |
| DTBench | — | 49.6% |
| BIG-Bench Hard | 55.1% | — |
| ForecastBench | — | 56.3 |
Math Gemma 7B leads
Gemma 7B: 31.2 (#228), Mixtral 8x7B: 18.8 (#289)
| Benchmark | Gemma 7B | Mixtral 8x7B |
|---|---|---|
| LMArena Math | 1066 | 1147 |
| GSM8K | 46.4% | 74.4% |
| Omni-MATH | — | 10.5% |
| MATH Level 5 | — | 10% |
Knowledge Gemma 7B leads
Gemma 7B: 27.3 (#252), Mixtral 8x7B: 11.0 (#301)
| Benchmark | Gemma 7B | Mixtral 8x7B |
|---|---|---|
| LMArena Expert | 1001 | 1088 |
| ARC (AI2) Challenge | 78.3% | 87.3% |
| MMLU | 66.1% | 70.6% |
| OpenBookQA | 78.6% | 85.8% |
| TriviaQA | 72.3% | 82.2% |
| GPQA Diamond | — | 30.6% |
| MMLU-Pro | — | 33.5% |
| GPQA (HELM) | — | 29.6% |
| BoolQ | 83.2% | — |
Multilingual Mixtral 8x7B leads
Gemma 7B: 25.1 (#287), Mixtral 8x7B: 29.6 (#266)
| Benchmark | Gemma 7B | Mixtral 8x7B |
|---|---|---|
| LMArena Non-English | 999 | 1077 |
| LMArena Chinese | 1035 | 1055 |
| LMArena French | 1025 | 1166 |
| LMArena Russian | 993 | 1090 |
| LMArena German | — | 1114 |
| LMArena Japanese | — | 931 |
| LMArena Korean | — | 968 |
| LMArena Spanish | — | 1111 |
Instruction Following Too close to call
Gemma 7B: 51.5 (#295), Mixtral 8x7B: 51.0 (#297)
| Benchmark | Gemma 7B | Mixtral 8x7B |
|---|---|---|
| LMArena Instruction Following | 1017 | 1109 |
| IFEval | — | 57.5% |
Long Context Mixtral 8x7B leads
Gemma 7B: 31.1 (#282), Mixtral 8x7B: 33.4 (#260)
| Benchmark | Gemma 7B | Mixtral 8x7B |
|---|---|---|
| LMArena Longer Query | 1022 | 1103 |
Writing & Preference Mixtral 8x7B leads
Gemma 7B: 27.1 (#302), Mixtral 8x7B: 34.2 (#270)
| Benchmark | Gemma 7B | Mixtral 8x7B |
|---|---|---|
| LMArena Text | 1056 | 1132 |
| LMArena Creative Writing | 1024 | 1109 |
| LMArena Multi-Turn | 963 | 1115 |
| WildBench | — | 67.3% |
Frequently asked questions
Is Gemma 7B better than Mixtral 8x7B?
Gemma 7B is the stronger model overall, scoring 30.0 to 27.1 on the Noometry Index.
Is Gemma 7B or Mixtral 8x7B better for coding?
Mixtral 8x7B scores higher on coding benchmarks: 32.8 versus 30.5 in the Noometry coding category.
How many benchmarks do Gemma 7B and Mixtral 8x7B share?
25 benchmarks have published results for both models. Gemma 7B has 27 scored results on Noometry and Mixtral 8x7B has 38.