Model comparison
Gemma 3 4B vs Mistral Large
Mistral Large is the stronger model overall, scoring 31.9 to 28.1 on the Noometry Index. Gemma 3 4B costs 60× less per token, which makes it the better buy when Mistral Large's lead doesn't matter for your workload.
Last verified . 20 shared benchmarks.
Google
Mistral AISummary
- They share 20 benchmarks with published results for both. Gemma 3 4B scores higher in 4 categories and Mistral Large in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Mistral Large leads 30.1 to 11.8.
- The biggest single-benchmark swing is GPQA Diamond: 23.2% for Gemma 3 4B and 51.3% for Mistral Large.
- Gemma 3 4B is cheaper at $0.04 / $0.08 per million input/output tokens, against $2 / $6 for Mistral Large.
Side by side
| Gemma 3 4B | Mistral Large | |
|---|---|---|
| Provider | Mistral AI | |
| Noometry Index | 28.1 | 31.9 |
| Released | 2025-03-12 | 2024-02-26 |
| Weights | Open | Open |
| Context window | 131K | 131K |
| Max output | 4K | 16K |
| Input $ / M tokens | $0.04 | $2 |
| Output $ / M tokens | $0.08 | $6 |
| Results tracked | 22 | 51 |
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), Mistral Large: 34.3 (#240)
| Benchmark | Gemma 3 4B | Mistral Large |
|---|---|---|
| LMArena Coding | 1230 | 1277 |
| SciCode | — | 36.2% |
| BigCodeBench Instruct | — | 30% |
| LiveBench Coding | — | 47.1% |
| BigCodeBench Complete | — | 38.3% |
| ALE-Bench | — | 264.7 |
| HumanEval+ | — | 62.2% |
| MBPP+ | — | 59.5% |
Agentic & Tool Use Mistral Large leads
Gemma 3 4B: 20.9 (#142), Mistral Large: 28.6 (#89)
| Benchmark | Gemma 3 4B | Mistral Large |
|---|---|---|
| Berkeley Function Calling Leaderboard | 19.6% | 38.4% |
Reasoning Mistral Large leads
Gemma 3 4B: 13.2 (#335), Mistral Large: 15.8 (#310)
| Benchmark | Gemma 3 4B | Mistral Large |
|---|---|---|
| LMArena Hard Prompts | 1253 | 1257 |
| DTBench | 50.9% | 65.1% |
| LMCA | 2.8% | 16.7% |
| Epoch Capabilities Index | 116.02 | 128.52 |
| SimpleBench | — | 22.5% |
| Kagi LLM Benchmark | 25.2% | — |
| CritPt | — | 0% |
| Chess Puzzles | 0% | — |
| LiveBench Reasoning | — | 43.5% |
| LiveBench Data Analysis | — | 50.1% |
| ForecastBench | — | 57.1 |
| LiveBench | — | 48.4% |
Math Mistral Large leads
Gemma 3 4B: 16.8 (#292), Mistral Large: 18.2 (#291)
| Benchmark | Gemma 3 4B | Mistral Large |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 7.5% | 8.5% |
| LMArena Math | 1239 | 1262 |
| Omni-MATH | — | 28.1% |
| LiveBench Math | — | 42.5% |
| MATH Level 5 | — | 50.3% |
| FrontierMath (Feb 2025 set) | — | 0.3% |
Knowledge Mistral Large leads
Gemma 3 4B: 11.8 (#299), Mistral Large: 30.1 (#230)
| Benchmark | Gemma 3 4B | Mistral Large |
|---|---|---|
| GPQA Diamond | 23.2% | 51.3% |
| Vectara Hallucination Rate | 6.4% | 4.5% |
| LMArena Expert | 1223 | 1232 |
| MMLU-Pro | — | 59.9% |
| Confabulations | — | 21.4% |
| GPQA (HELM) | — | 43.5% |
| MMLU | — | 80% |
Multilingual Gemma 3 4B leads
Gemma 3 4B: 42.5 (#194), Mistral Large: 40.0 (#219)
| Benchmark | Gemma 3 4B | Mistral Large |
|---|---|---|
| LMArena Non-English | 1273 | 1237 |
| LMArena German | 1281 | 1254 |
| LMArena Russian | 1294 | 1257 |
| LMArena Chinese | — | 1240 |
| LMArena French | — | 1325 |
| LMArena Japanese | — | 1188 |
| LMArena Korean | — | 1202 |
| LMArena Spanish | — | 1268 |
Instruction Following Mistral Large leads
Gemma 3 4B: 65.2 (#225), Mistral Large: 67.9 (#191)
| Benchmark | Gemma 3 4B | Mistral Large |
|---|---|---|
| LMArena Instruction Following | 1239 | 1249 |
| LiveBench Instruction Following | — | 67.9% |
| IFEval | — | 87.7% |
Long Context Too close to call
Gemma 3 4B: 38.7 (#194), Mistral Large: 38.3 (#199)
| Benchmark | Gemma 3 4B | Mistral Large |
|---|---|---|
| LMArena Longer Query | 1273 | 1261 |
Writing & Preference Gemma 3 4B leads
Gemma 3 4B: 42.0 (#239), Mistral Large: 40.7 (#242)
| Benchmark | Gemma 3 4B | Mistral Large |
|---|---|---|
| LMArena Text | 1291 | 1266 |
| LMArena Creative Writing | 1271 | 1243 |
| EQ-Bench Creative Writing | 1068 | 985 |
| LMArena Multi-Turn | 1255 | 1260 |
| Short-Story Creative Writing | — | 69% |
| WildBench | — | 80.1% |
| LiveBench Language | — | 39.4% |
Frequently asked questions
Is Gemma 3 4B better than Mistral Large?
Mistral Large is the stronger model overall, scoring 31.9 to 28.1 on the Noometry Index. Gemma 3 4B costs 60× less per token, which makes it the better buy when Mistral Large's lead doesn't matter for your workload.
Which is cheaper, Gemma 3 4B or Mistral Large?
Gemma 3 4B is cheaper. It lists at $0.04 per million input tokens and $0.08 per million output tokens; Mistral Large lists at $2 and $6.
Is Gemma 3 4B or Mistral Large better for coding?
Gemma 3 4B scores higher on coding benchmarks: 35.9 versus 34.3 in the Noometry coding category.
Which has the bigger context window?
Both accept 131K tokens.
How many benchmarks do Gemma 3 4B and Mistral Large share?
20 benchmarks have published results for both models. Gemma 3 4B has 22 scored results on Noometry and Mistral Large has 51.