Model comparison
Gemma 3 27B vs Mistral Large
Mistral Large is the stronger model overall, scoring 31.9 to 30.8 on the Noometry Index. Gemma 3 27B costs 30× less per token, which makes it the better buy when Mistral Large's lead doesn't matter for your workload.
Last verified . 37 shared benchmarks.
Google
Mistral AISummary
- They share 37 benchmarks with published results for both. Gemma 3 27B scores higher in 5 categories and Mistral Large in 4 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Gemma 3 27B leads 52.5 to 40.7.
- The biggest single-benchmark swing is MATH Level 5: 74% for Gemma 3 27B and 50.3% for Mistral Large.
- Gemma 3 27B is cheaper at $0.08 / $0.16 per million input/output tokens, against $2 / $6 for Mistral Large.
Side by side
| Gemma 3 27B | Mistral Large | |
|---|---|---|
| Provider | Mistral AI | |
| Noometry Index | 30.8 | 31.9 |
| Released | 2025-03-11 | 2024-02-26 |
| Weights | Open | Open |
| Context window | 131K | 131K |
| Max output | 8K | 16K |
| Input $ / M tokens | $0.08 | $2 |
| Output $ / M tokens | $0.16 | $6 |
| Results tracked | 43 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mistral Large leads
Gemma 3 27B: 22.5 (#334), Mistral Large: 34.3 (#240)
| Benchmark | Gemma 3 27B | Mistral Large |
|---|---|---|
| SciCode | 21.2% | 36.2% |
| LiveBench Coding | 39.9% | 47.1% |
| LMArena Coding | 1322 | 1277 |
| Aider Polyglot | 4.9% | — |
| BigCodeBench Instruct | — | 30% |
| BigCodeBench Complete | — | 38.3% |
| ALE-Bench | — | 264.7 |
| HumanEval+ | — | 62.2% |
| MBPP+ | — | 59.5% |
Agentic & Tool Use Mistral Large leads
Gemma 3 27B: 25.1 (#110), Mistral Large: 28.6 (#89)
| Benchmark | Gemma 3 27B | Mistral Large |
|---|---|---|
| Berkeley Function Calling Leaderboard | 29.5% | 38.4% |
Reasoning Too close to call
Gemma 3 27B: 16.7 (#301), Mistral Large: 15.8 (#310)
| Benchmark | Gemma 3 27B | Mistral Large |
|---|---|---|
| CritPt | 0% | 0% |
| LiveBench Reasoning | 43.8% | 43.5% |
| LMArena Hard Prompts | 1340 | 1257 |
| DTBench | 52.5% | 65.1% |
| LiveBench Data Analysis | 51.5% | 50.1% |
| LMCA | 12.3% | 16.7% |
| Epoch Capabilities Index | 130.04 | 128.52 |
| LiveBench | 50% | 48.4% |
| SimpleBench | — | 22.5% |
| Kagi LLM Benchmark | 40.4% | — |
| Chess Puzzles | 0% | — |
| ForecastBench | — | 57.1 |
Math Gemma 3 27B leads
Gemma 3 27B: 25.9 (#265), Mistral Large: 18.2 (#291)
| Benchmark | Gemma 3 27B | Mistral Large |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 22.5% | 8.5% |
| LiveBench Math | 55.4% | 42.5% |
| LMArena Math | 1312 | 1262 |
| MATH Level 5 | 74% | 50.3% |
| Omni-MATH | — | 28.1% |
| FrontierMath (Feb 2025 set) | — | 0.3% |
Knowledge Mistral Large leads
Gemma 3 27B: 25.5 (#261), Mistral Large: 30.1 (#230)
| Benchmark | Gemma 3 27B | Mistral Large |
|---|---|---|
| GPQA Diamond | 47.7% | 51.3% |
| Confabulations | 40.3% | 21.4% |
| Vectara Hallucination Rate | 7.4% | 4.5% |
| LMArena Expert | 1304 | 1232 |
| MMLU-Pro | — | 59.9% |
| GPQA (HELM) | — | 43.5% |
| MMLU | — | 80% |
Multimodal Not comparable
Gemma 3 27B: 32.6 (#100), Mistral Large: —
| Benchmark | Gemma 3 27B | Mistral Large |
|---|---|---|
| LMArena Vision | 1164 | — |
| GeoBench | 52% | — |
Multilingual Gemma 3 27B leads
Gemma 3 27B: 46.9 (#155), Mistral Large: 40.0 (#219)
| Benchmark | Gemma 3 27B | Mistral Large |
|---|---|---|
| LMArena Non-English | 1334 | 1237 |
| LMArena Chinese | 1346 | 1240 |
| LMArena French | 1368 | 1325 |
| LMArena German | 1362 | 1254 |
| LMArena Japanese | 1287 | 1188 |
| LMArena Korean | 1308 | 1202 |
| LMArena Russian | 1349 | 1257 |
| LMArena Spanish | 1349 | 1268 |
Instruction Following Gemma 3 27B leads
Gemma 3 27B: 70.6 (#160), Mistral Large: 67.9 (#191)
| Benchmark | Gemma 3 27B | Mistral Large |
|---|---|---|
| LiveBench Instruction Following | 74.9% | 67.9% |
| LMArena Instruction Following | 1321 | 1249 |
| IFEval | — | 87.7% |
Long Context Mistral Large leads
Gemma 3 27B: 27.6 (#293), Mistral Large: 38.3 (#199)
| Benchmark | Gemma 3 27B | Mistral Large |
|---|---|---|
| LMArena Longer Query | 1333 | 1261 |
| Fiction.LiveBench | 33.3% | — |
Writing & Preference Gemma 3 27B leads
Gemma 3 27B: 52.5 (#168), Mistral Large: 40.7 (#242)
| Benchmark | Gemma 3 27B | Mistral Large |
|---|---|---|
| LMArena Text | 1358 | 1266 |
| LMArena Creative Writing | 1346 | 1243 |
| Short-Story Creative Writing | 79.9% | 69% |
| EQ-Bench Creative Writing | 1266 | 985 |
| LMArena Multi-Turn | 1345 | 1260 |
| LiveBench Language | 34.6% | 39.4% |
| WildBench | — | 80.1% |
Frequently asked questions
Is Gemma 3 27B better than Mistral Large?
Mistral Large is the stronger model overall, scoring 31.9 to 30.8 on the Noometry Index. Gemma 3 27B costs 30× 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 27B or Mistral Large?
Gemma 3 27B is cheaper. It lists at $0.08 per million input tokens and $0.16 per million output tokens; Mistral Large lists at $2 and $6.
Is Gemma 3 27B or Mistral Large better for coding?
Mistral Large scores higher on coding benchmarks: 34.3 versus 22.5 in the Noometry coding category.
Which has the bigger context window?
Both accept 131K tokens.
How many benchmarks do Gemma 3 27B and Mistral Large share?
37 benchmarks have published results for both models. Gemma 3 27B has 43 scored results on Noometry and Mistral Large has 51.