Model comparison
Gemma 7B vs Mixtral 8x22B
Gemma 7B is the stronger model overall, scoring 29.5 to 26.7 on the Noometry Index.
Last verified . 17 shared benchmarks.
Google
Mistral AISummary
- They share 17 benchmarks with published results for both. Gemma 7B scores higher in 3 categories and Mixtral 8x22B in 5 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Gemma 7B leads 26.8 to 15.1.
Side by side
| Gemma 7B | Mixtral 8x22B | |
|---|---|---|
| Provider | Mistral AI | |
| Noometry Index | 29.5 | 26.7 |
| Released | 2024-02-21 | 2024-04-17 |
| Weights | Open | Open |
| Context window | — | 64K |
| Max output | — | 64K |
| Input $ / M tokens | — | $2 |
| Output $ / M tokens | — | $6 |
| Results tracked | 27 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 7B leads
Gemma 7B: 29.9 (#290), Mixtral 8x22B: 23.8 (#324)
| Benchmark | Gemma 7B | Mixtral 8x22B |
|---|---|---|
| LMArena Coding | 1048 | 1166 |
| HumanEval+ | 28.7% | 72% |
| MBPP+ | 43.4% | 64.3% |
| WeirdML | — | 3.2% |
| BigCodeBench Instruct | — | 40.6% |
| BigCodeBench Complete | — | 50.2% |
Agentic & Tool Use Not comparable
Gemma 7B: —, Mixtral 8x22B: 22.8 (#119)
| Benchmark | Gemma 7B | Mixtral 8x22B |
|---|---|---|
| Cybench | — | 7.5% |
Reasoning Too close to call
Gemma 7B: 19.0 (#247), Mixtral 8x22B: 19.1 (#246)
| Benchmark | Gemma 7B | Mixtral 8x22B |
|---|---|---|
| LMArena Hard Prompts | 1042 | 1150 |
| Epoch Capabilities Index | 111.74 | 122 |
| DTBench | — | 55.1% |
| Adversarial NLI | 48.7% | — |
| BIG-Bench Hard | 55.1% | — |
| ForecastBench | — | 56.3 |
| HellaSwag | 82.2% | — |
| PIQA | 81.2% | — |
| WinoGrande | 79% | — |
Math Gemma 7B leads
Gemma 7B: 28.7 (#225), Mixtral 8x22B: 21.2 (#271)
| Benchmark | Gemma 7B | Mixtral 8x22B |
|---|---|---|
| LMArena Math | 1066 | 1184 |
| Omni-MATH | — | 16.3% |
| MATH Level 5 | — | 24.2% |
| GSM8K | 46.4% | — |
Knowledge Gemma 7B leads
Gemma 7B: 26.8 (#248), Mixtral 8x22B: 15.1 (#288)
| Benchmark | Gemma 7B | Mixtral 8x22B |
|---|---|---|
| LMArena Expert | 1002 | 1113 |
| MMLU | 66.1% | 77.8% |
| GPQA Diamond | — | 34.1% |
| MMLU-Pro | — | 46% |
| GPQA (HELM) | — | 33.4% |
| ARC (AI2) Challenge | 78.3% | — |
| BoolQ | 83.2% | — |
| OpenBookQA | 78.6% | — |
| TriviaQA | 72.3% | — |
Multilingual Mixtral 8x22B leads
Gemma 7B: 25.3 (#282), Mixtral 8x22B: 32.9 (#250)
| Benchmark | Gemma 7B | Mixtral 8x22B |
|---|---|---|
| LMArena Non-English | 999 | 1128 |
| LMArena Chinese | 1035 | 1116 |
| LMArena French | 1024 | 1165 |
| LMArena Russian | 993 | 1158 |
| LMArena German | — | 1141 |
| LMArena Japanese | — | 1037 |
| LMArena Korean | — | 1056 |
| LMArena Spanish | — | 1150 |
Instruction Following Mixtral 8x22B leads
Gemma 7B: 51.7 (#290), Mixtral 8x22B: 57.9 (#261)
| Benchmark | Gemma 7B | Mixtral 8x22B |
|---|---|---|
| LMArena Instruction Following | 1017 | 1147 |
| IFEval | — | 72.4% |
Long Context Mixtral 8x22B leads
Gemma 7B: 31.1 (#277), Mixtral 8x22B: 34.6 (#242)
| Benchmark | Gemma 7B | Mixtral 8x22B |
|---|---|---|
| LMArena Longer Query | 1022 | 1144 |
Writing & Preference Mixtral 8x22B leads
Gemma 7B: 27.2 (#297), Mixtral 8x22B: 37.1 (#257)
| Benchmark | Gemma 7B | Mixtral 8x22B |
|---|---|---|
| LMArena Text | 1056 | 1162 |
| LMArena Creative Writing | 1024 | 1141 |
| LMArena Multi-Turn | 963 | 1130 |
| WildBench | — | 71.1% |
Frequently asked questions
Is Gemma 7B better than Mixtral 8x22B?
Gemma 7B is the stronger model overall, scoring 29.5 to 26.7 on the Noometry Index.
Is Gemma 7B or Mixtral 8x22B better for coding?
Gemma 7B scores higher on coding benchmarks: 29.9 versus 23.8 in the Noometry coding category.
How many benchmarks do Gemma 7B and Mixtral 8x22B share?
17 benchmarks have published results for both models. Gemma 7B has 27 scored results on Noometry and Mixtral 8x22B has 34.