Model comparison
Gemma 3 27B vs Mistral Medium
Mistral Medium is the stronger model overall, scoring 36.3 to 30.8 on the Noometry Index. Gemma 3 27B costs 30× less per token, which makes it the better buy when Mistral Medium's lead doesn't matter for your workload.
Last verified . 29 shared benchmarks.
Google
Mistral AISummary
- They share 29 benchmarks with published results for both. Gemma 3 27B scores higher in 1 category and Mistral Medium in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in long context, where Mistral Medium leads 42.9 to 27.6.
- The biggest single-benchmark swing is DTBench: 52.5% for Gemma 3 27B and 75.5% for Mistral Medium.
- Gemma 3 27B is cheaper at $0.08 / $0.16 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 27B | Mistral Medium | |
|---|---|---|
| Provider | Mistral AI | |
| Noometry Index | 30.8 | 36.3 |
| Released | 2025-03-11 | 2023-12-11 |
| Weights | Open | Open |
| Context window | 131K | 262K |
| Max output | 8K | 262K |
| Input $ / M tokens | $0.08 | $1.50 |
| Output $ / M tokens | $0.16 | $7.50 |
| Results tracked | 43 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mistral Medium leads
Gemma 3 27B: 22.5 (#334), Mistral Medium: 34.2 (#243)
| Benchmark | Gemma 3 27B | Mistral Medium |
|---|---|---|
| SciCode | 21.2% | 40.2% |
| LMArena Coding | 1322 | 1434 |
| FrontierCode | — | 8% |
| Aider Polyglot | 4.9% | — |
| WeirdML | — | 43.7% |
| LiveBench Coding | 39.9% | — |
| ALE-Bench | — | 763.98 |
Agentic & Tool Use Mistral Medium leads
Gemma 3 27B: 25.1 (#110), Mistral Medium: 28.3 (#90)
| Benchmark | Gemma 3 27B | Mistral Medium |
|---|---|---|
| Berkeley Function Calling Leaderboard | 29.5% | 37.7% |
Reasoning Mistral Medium leads
Gemma 3 27B: 16.7 (#301), Mistral Medium: 24.0 (#167)
| Benchmark | Gemma 3 27B | Mistral Medium |
|---|---|---|
| Kagi LLM Benchmark | 40.4% | 50% |
| CritPt | 0% | 0% |
| LMArena Hard Prompts | 1340 | 1426 |
| DTBench | 52.5% | 75.5% |
| LMCA | 12.3% | 26.1% |
| Chess Puzzles | 0% | — |
| LiveBench Reasoning | 43.8% | — |
| LiveBench Data Analysis | 51.5% | — |
| Surface Evolver Bench | — | 26.9% |
| Epoch Capabilities Index | 130.04 | — |
| LiveBench | 50% | — |
Math Mistral Medium leads
Gemma 3 27B: 25.9 (#265), Mistral Medium: 28.1 (#245)
| Benchmark | Gemma 3 27B | Mistral Medium |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 22.5% | 32.2% |
| LMArena Math | 1312 | 1408 |
| MATH Level 5 | 74% | 81.6% |
| ProofBench | — | 9% |
| LiveBench Math | 55.4% | — |
| FrontierMath (Feb 2025 set) | — | 0.3% |
Knowledge Too close to call
Gemma 3 27B: 25.5 (#261), Mistral Medium: 25.0 (#265)
| Benchmark | Gemma 3 27B | Mistral Medium |
|---|---|---|
| GPQA Diamond | 47.7% | 59.5% |
| Vectara Hallucination Rate | 7.4% | 22.7% |
| LMArena Expert | 1304 | 1408 |
| Humanity's Last Exam | — | 4.5% |
| Confabulations | 40.3% | — |
Multimodal Mistral Medium leads
Gemma 3 27B: 32.6 (#100), Mistral Medium: 35.3 (#88)
| Benchmark | Gemma 3 27B | Mistral Medium |
|---|---|---|
| LMArena Vision | 1164 | 1172 |
| GeoBench | 52% | — |
Multilingual Mistral Medium leads
Gemma 3 27B: 46.9 (#155), Mistral Medium: 52.1 (#91)
| Benchmark | Gemma 3 27B | Mistral Medium |
|---|---|---|
| LMArena Non-English | 1334 | 1408 |
| LMArena Chinese | 1346 | 1447 |
| LMArena French | 1368 | 1459 |
| LMArena German | 1362 | 1432 |
| LMArena Japanese | 1287 | 1378 |
| LMArena Korean | 1308 | 1380 |
| LMArena Russian | 1349 | 1411 |
| LMArena Spanish | 1349 | 1433 |
Instruction Following Mistral Medium leads
Gemma 3 27B: 70.6 (#160), Mistral Medium: 73.7 (#116)
| Benchmark | Gemma 3 27B | Mistral Medium |
|---|---|---|
| LMArena Instruction Following | 1321 | 1398 |
| LiveBench Instruction Following | 74.9% | — |
Long Context Mistral Medium leads
Gemma 3 27B: 27.6 (#293), Mistral Medium: 42.9 (#114)
| Benchmark | Gemma 3 27B | Mistral Medium |
|---|---|---|
| LMArena Longer Query | 1333 | 1406 |
| Fiction.LiveBench | 33.3% | — |
Writing & Preference Mistral Medium leads
Gemma 3 27B: 52.5 (#168), Mistral Medium: 60.0 (#103)
| Benchmark | Gemma 3 27B | Mistral Medium |
|---|---|---|
| LMArena Text | 1358 | 1424 |
| LMArena Creative Writing | 1346 | 1391 |
| Short-Story Creative Writing | 79.9% | 77.3% |
| LMArena Multi-Turn | 1345 | 1418 |
| EQ-Bench Creative Writing | 1266 | — |
| LiveBench Language | 34.6% | — |
Frequently asked questions
Is Gemma 3 27B better than Mistral Medium?
Mistral Medium is the stronger model overall, scoring 36.3 to 30.8 on the Noometry Index. Gemma 3 27B costs 30× 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 27B or Mistral Medium?
Gemma 3 27B is cheaper. It lists at $0.08 per million input tokens and $0.16 per million output tokens; Mistral Medium lists at $1.50 and $7.50.
Is Gemma 3 27B or Mistral Medium better for coding?
Mistral Medium scores higher on coding benchmarks: 34.2 versus 22.5 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 27B and Mistral Medium share?
29 benchmarks have published results for both models. Gemma 3 27B has 43 scored results on Noometry and Mistral Medium has 36.