Model comparison
Gemma 3 4B vs Mixtral 8x22B
Gemma 3 4B and Mixtral 8x22B score almost the same on the Noometry Index (28.1 vs 27.1), so choose on price, context window or the category you care about most.
Last verified . 15 shared benchmarks.
Google
Mistral AISummary
- They share 15 benchmarks with published results for both. Gemma 3 4B scores higher in 5 categories and Mixtral 8x22B in 4 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in coding, where Gemma 3 4B leads 35.9 to 24.2.
- The biggest single-benchmark swing is GPQA Diamond: 23.2% for Gemma 3 4B and 34.1% for Mixtral 8x22B.
- Gemma 3 4B is cheaper at $0.04 / $0.08 per million input/output tokens, against $2 / $6 for Mixtral 8x22B.
- Gemma 3 4B accepts more context: 131K tokens versus 64K.
Side by side
| Gemma 3 4B | Mixtral 8x22B | |
|---|---|---|
| Provider | Mistral AI | |
| Noometry Index | 28.1 | 27.1 |
| Released | 2025-03-12 | 2024-04-17 |
| Weights | Open | Open |
| Context window | 131K | 64K |
| Max output | 4K | 64K |
| Input $ / M tokens | $0.04 | $2 |
| Output $ / M tokens | $0.08 | $6 |
| Results tracked | 22 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 3 4B leads
Gemma 3 4B: 35.9 (#215), Mixtral 8x22B: 24.2 (#329)
| Benchmark | Gemma 3 4B | Mixtral 8x22B |
|---|---|---|
| LMArena Coding | 1230 | 1166 |
| WeirdML | — | 3.2% |
| BigCodeBench Instruct | — | 40.6% |
| BigCodeBench Complete | — | 50.2% |
| HumanEval+ | — | 72% |
| MBPP+ | — | 64.3% |
Agentic & Tool Use Mixtral 8x22B leads
Gemma 3 4B: 20.9 (#142), Mixtral 8x22B: 23.1 (#127)
| Benchmark | Gemma 3 4B | Mixtral 8x22B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 19.6% | — |
| Cybench | — | 7.5% |
Reasoning Mixtral 8x22B leads
Gemma 3 4B: 13.2 (#335), Mixtral 8x22B: 19.9 (#248)
| Benchmark | Gemma 3 4B | Mixtral 8x22B |
|---|---|---|
| LMArena Hard Prompts | 1253 | 1150 |
| DTBench | 50.9% | 55.1% |
| Epoch Capabilities Index | 116.02 | 122.03 |
| Kagi LLM Benchmark | 25.2% | — |
| Chess Puzzles | 0% | — |
| LMCA | 2.8% | — |
| ForecastBench | — | 56.3 |
Math Mixtral 8x22B leads
Gemma 3 4B: 16.8 (#292), Mixtral 8x22B: 22.9 (#275)
| Benchmark | Gemma 3 4B | Mixtral 8x22B |
|---|---|---|
| LMArena Math | 1239 | 1184 |
| OTIS Mock AIME 2024-2025 | 7.5% | — |
| Omni-MATH | — | 16.3% |
| MATH Level 5 | — | 24.2% |
Knowledge Mixtral 8x22B leads
Gemma 3 4B: 11.8 (#299), Mixtral 8x22B: 15.1 (#293)
| Benchmark | Gemma 3 4B | Mixtral 8x22B |
|---|---|---|
| GPQA Diamond | 23.2% | 34.1% |
| LMArena Expert | 1223 | 1113 |
| MMLU-Pro | — | 46% |
| Vectara Hallucination Rate | 6.4% | — |
| GPQA (HELM) | — | 33.4% |
| MMLU | — | 77.8% |
Multilingual Gemma 3 4B leads
Gemma 3 4B: 42.5 (#194), Mixtral 8x22B: 32.8 (#255)
| Benchmark | Gemma 3 4B | Mixtral 8x22B |
|---|---|---|
| LMArena Non-English | 1273 | 1128 |
| LMArena German | 1281 | 1141 |
| LMArena Russian | 1294 | 1158 |
| LMArena Chinese | — | 1116 |
| LMArena French | — | 1166 |
| LMArena Japanese | — | 1037 |
| LMArena Korean | — | 1057 |
| LMArena Spanish | — | 1151 |
Instruction Following Gemma 3 4B leads
Gemma 3 4B: 65.2 (#225), Mixtral 8x22B: 57.7 (#266)
| Benchmark | Gemma 3 4B | Mixtral 8x22B |
|---|---|---|
| LMArena Instruction Following | 1239 | 1147 |
| IFEval | — | 72.4% |
Long Context Gemma 3 4B leads
Gemma 3 4B: 38.7 (#194), Mixtral 8x22B: 34.7 (#247)
| Benchmark | Gemma 3 4B | Mixtral 8x22B |
|---|---|---|
| LMArena Longer Query | 1273 | 1144 |
Writing & Preference Gemma 3 4B leads
Gemma 3 4B: 42.0 (#239), Mixtral 8x22B: 36.9 (#262)
| Benchmark | Gemma 3 4B | Mixtral 8x22B |
|---|---|---|
| LMArena Text | 1291 | 1162 |
| LMArena Creative Writing | 1271 | 1141 |
| LMArena Multi-Turn | 1255 | 1130 |
| EQ-Bench Creative Writing | 1068 | — |
| WildBench | — | 71.1% |
Frequently asked questions
Is Gemma 3 4B better than Mixtral 8x22B?
Gemma 3 4B and Mixtral 8x22B score almost the same on the Noometry Index (28.1 vs 27.1), so choose on price, context window or the category you care about most.
Which is cheaper, Gemma 3 4B or Mixtral 8x22B?
Gemma 3 4B is cheaper. It lists at $0.04 per million input tokens and $0.08 per million output tokens; Mixtral 8x22B lists at $2 and $6.
Is Gemma 3 4B or Mixtral 8x22B better for coding?
Gemma 3 4B scores higher on coding benchmarks: 35.9 versus 24.2 in the Noometry coding category.
Which has the bigger context window?
Gemma 3 4B does, with 131K tokens against 64K.
How many benchmarks do Gemma 3 4B and Mixtral 8x22B share?
15 benchmarks have published results for both models. Gemma 3 4B has 22 scored results on Noometry and Mixtral 8x22B has 34.