Model comparison
Gemma 3 4B vs Mistral Medium
Mistral Medium is the stronger model overall, scoring 36.3 to 28.1 on the Noometry Index. Gemma 3 4B costs 60× less per token, which makes it the better buy when Mistral Medium's lead doesn't matter for your workload.
Last verified . 19 shared benchmarks.
Google
Mistral AISummary
- They share 19 benchmarks with published results for both. Gemma 3 4B scores higher in 1 category and Mistral Medium in 8 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Mistral Medium leads 60.0 to 42.0.
- The biggest single-benchmark swing is GPQA Diamond: 23.2% for Gemma 3 4B and 59.5% for Mistral Medium.
- Gemma 3 4B is cheaper at $0.04 / $0.08 per million input/output tokens, against $1.50 / $7.50 for Mistral Medium.
- Mistral Medium accepts more context: 262K tokens versus 131K.
Side by side
| Gemma 3 4B | Mistral Medium | |
|---|---|---|
| Provider | Mistral AI | |
| Noometry Index | 28.1 | 36.3 |
| Released | 2025-03-12 | 2023-12-11 |
| Weights | Open | Open |
| Context window | 131K | 262K |
| Max output | 4K | 262K |
| Input $ / M tokens | $0.04 | $1.50 |
| Output $ / M tokens | $0.08 | $7.50 |
| Results tracked | 22 | 36 |
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 Medium: 34.2 (#243)
| Benchmark | Gemma 3 4B | Mistral Medium |
|---|---|---|
| LMArena Coding | 1230 | 1434 |
| FrontierCode | — | 8% |
| SciCode | — | 40.2% |
| WeirdML | — | 43.7% |
| ALE-Bench | — | 763.98 |
Agentic & Tool Use Mistral Medium leads
Gemma 3 4B: 20.9 (#142), Mistral Medium: 28.3 (#90)
| Benchmark | Gemma 3 4B | Mistral Medium |
|---|---|---|
| Berkeley Function Calling Leaderboard | 19.6% | 37.7% |
Reasoning Mistral Medium leads
Gemma 3 4B: 13.2 (#335), Mistral Medium: 24.0 (#167)
| Benchmark | Gemma 3 4B | Mistral Medium |
|---|---|---|
| Kagi LLM Benchmark | 25.2% | 50% |
| LMArena Hard Prompts | 1253 | 1426 |
| DTBench | 50.9% | 75.5% |
| LMCA | 2.8% | 26.1% |
| CritPt | — | 0% |
| Chess Puzzles | 0% | — |
| Surface Evolver Bench | — | 26.9% |
| Epoch Capabilities Index | 116.02 | — |
Math Mistral Medium leads
Gemma 3 4B: 16.8 (#292), Mistral Medium: 28.1 (#245)
| Benchmark | Gemma 3 4B | Mistral Medium |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 7.5% | 32.2% |
| LMArena Math | 1239 | 1408 |
| ProofBench | — | 9% |
| MATH Level 5 | — | 81.6% |
| FrontierMath (Feb 2025 set) | — | 0.3% |
Knowledge Mistral Medium leads
Gemma 3 4B: 11.8 (#299), Mistral Medium: 25.0 (#265)
| Benchmark | Gemma 3 4B | Mistral Medium |
|---|---|---|
| GPQA Diamond | 23.2% | 59.5% |
| Vectara Hallucination Rate | 6.4% | 22.7% |
| LMArena Expert | 1223 | 1408 |
| Humanity's Last Exam | — | 4.5% |
Multimodal Not comparable
Gemma 3 4B: —, Mistral Medium: 35.3 (#88)
| Benchmark | Gemma 3 4B | Mistral Medium |
|---|---|---|
| LMArena Vision | — | 1172 |
Multilingual Mistral Medium leads
Gemma 3 4B: 42.5 (#194), Mistral Medium: 52.1 (#91)
| Benchmark | Gemma 3 4B | Mistral Medium |
|---|---|---|
| LMArena Non-English | 1273 | 1408 |
| LMArena German | 1281 | 1432 |
| LMArena Russian | 1294 | 1411 |
| LMArena Chinese | — | 1447 |
| LMArena French | — | 1459 |
| LMArena Japanese | — | 1378 |
| LMArena Korean | — | 1380 |
| LMArena Spanish | — | 1433 |
Instruction Following Mistral Medium leads
Gemma 3 4B: 65.2 (#225), Mistral Medium: 73.7 (#116)
| Benchmark | Gemma 3 4B | Mistral Medium |
|---|---|---|
| LMArena Instruction Following | 1239 | 1398 |
Long Context Mistral Medium leads
Gemma 3 4B: 38.7 (#194), Mistral Medium: 42.9 (#114)
| Benchmark | Gemma 3 4B | Mistral Medium |
|---|---|---|
| LMArena Longer Query | 1273 | 1406 |
Writing & Preference Mistral Medium leads
Gemma 3 4B: 42.0 (#239), Mistral Medium: 60.0 (#103)
| Benchmark | Gemma 3 4B | Mistral Medium |
|---|---|---|
| LMArena Text | 1291 | 1424 |
| LMArena Creative Writing | 1271 | 1391 |
| LMArena Multi-Turn | 1255 | 1418 |
| Short-Story Creative Writing | — | 77.3% |
| EQ-Bench Creative Writing | 1068 | — |
Frequently asked questions
Is Gemma 3 4B better than Mistral Medium?
Mistral Medium is the stronger model overall, scoring 36.3 to 28.1 on the Noometry Index. Gemma 3 4B costs 60× less per token, which makes it the better buy when Mistral Medium's lead doesn't matter for your workload.
Which is cheaper, Gemma 3 4B or Mistral Medium?
Gemma 3 4B is cheaper. It lists at $0.04 per million input tokens and $0.08 per million output tokens; Mistral Medium lists at $1.50 and $7.50.
Is Gemma 3 4B or Mistral Medium better for coding?
Gemma 3 4B scores higher on coding benchmarks: 35.9 versus 34.2 in the Noometry coding category.
Which has the bigger context window?
Mistral Medium does, with 262K tokens against 131K.
How many benchmarks do Gemma 3 4B and Mistral Medium share?
19 benchmarks have published results for both models. Gemma 3 4B has 22 scored results on Noometry and Mistral Medium has 36.