Model comparison
Gemma 2 27B vs Mistral Large
Mistral Large is the stronger model overall, scoring 31.9 to 29.4 on the Noometry Index. Gemma 2 27B costs 4.6× less per token, which makes it the better buy when Mistral Large's lead doesn't matter for your workload.
Last verified . 34 shared benchmarks.
Google
Mistral AISummary
- They share 34 benchmarks with published results for both. Gemma 2 27B scores higher in 1 category and Mistral Large in 7 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Mistral Large leads 30.1 to 19.0.
- The biggest single-benchmark swing is MATH Level 5: 27.9% for Gemma 2 27B and 50.3% for Mistral Large.
- Gemma 2 27B is cheaper at $0.65 / $0.65 per million input/output tokens, against $2 / $6 for Mistral Large.
- Mistral Large accepts more context: 131K tokens versus 8K.
Side by side
| Gemma 2 27B | Mistral Large | |
|---|---|---|
| Provider | Mistral AI | |
| Noometry Index | 29.4 | 31.9 |
| Released | 2024-06-24 | 2024-02-26 |
| Weights | Open | Open |
| Context window | 8K | 131K |
| Max output | 2K | 16K |
| Input $ / M tokens | $0.65 | $2 |
| Output $ / M tokens | $0.65 | $6 |
| Results tracked | 34 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Gemma 2 27B: 34.1 (#246), Mistral Large: 34.3 (#240)
| Benchmark | Gemma 2 27B | Mistral Large |
|---|---|---|
| BigCodeBench Instruct | 42.8% | 30% |
| LiveBench Coding | 36% | 47.1% |
| LMArena Coding | 1211 | 1277 |
| BigCodeBench Complete | 52.5% | 38.3% |
| SciCode | — | 36.2% |
| ALE-Bench | — | 264.7 |
| HumanEval+ | — | 62.2% |
| MBPP+ | — | 59.5% |
Agentic & Tool Use Not comparable
Gemma 2 27B: —, Mistral Large: 28.6 (#89)
| Benchmark | Gemma 2 27B | Mistral Large |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 38.4% |
Reasoning Too close to call
Gemma 2 27B: 15.3 (#315), Mistral Large: 15.8 (#310)
| Benchmark | Gemma 2 27B | Mistral Large |
|---|---|---|
| LiveBench Reasoning | 28.1% | 43.5% |
| LMArena Hard Prompts | 1198 | 1257 |
| DTBench | 48% | 65.1% |
| LiveBench Data Analysis | 47.9% | 50.1% |
| LMCA | 7.1% | 16.7% |
| Epoch Capabilities Index | 122.08 | 128.52 |
| LiveBench | 38.2% | 48.4% |
| SimpleBench | — | 22.5% |
| CritPt | — | 0% |
| ForecastBench | — | 57.1 |
Math Mistral Large leads
Gemma 2 27B: 10.7 (#311), Mistral Large: 18.2 (#291)
| Benchmark | Gemma 2 27B | Mistral Large |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.4% | 8.5% |
| LiveBench Math | 26.5% | 42.5% |
| LMArena Math | 1212 | 1262 |
| MATH Level 5 | 27.9% | 50.3% |
| Omni-MATH | — | 28.1% |
| FrontierMath (Feb 2025 set) | — | 0.3% |
Knowledge Mistral Large leads
Gemma 2 27B: 19.0 (#280), Mistral Large: 30.1 (#230)
| Benchmark | Gemma 2 27B | Mistral Large |
|---|---|---|
| GPQA Diamond | 36.5% | 51.3% |
| Confabulations | 27.1% | 21.4% |
| LMArena Expert | 1172 | 1232 |
| MMLU | 75.7% | 80% |
| MMLU-Pro | — | 59.9% |
| Vectara Hallucination Rate | — | 4.5% |
| GPQA (HELM) | — | 43.5% |
Multilingual Mistral Large leads
Gemma 2 27B: 38.6 (#226), Mistral Large: 40.0 (#219)
| Benchmark | Gemma 2 27B | Mistral Large |
|---|---|---|
| LMArena Non-English | 1217 | 1237 |
| LMArena Chinese | 1221 | 1240 |
| LMArena French | 1247 | 1325 |
| LMArena German | 1209 | 1254 |
| LMArena Japanese | 1175 | 1188 |
| LMArena Korean | 1174 | 1202 |
| LMArena Russian | 1234 | 1257 |
| LMArena Spanish | 1228 | 1268 |
Instruction Following Mistral Large leads
Gemma 2 27B: 60.5 (#249), Mistral Large: 67.9 (#191)
| Benchmark | Gemma 2 27B | Mistral Large |
|---|---|---|
| LiveBench Instruction Following | 58.1% | 67.9% |
| LMArena Instruction Following | 1206 | 1249 |
| IFEval | — | 87.7% |
Long Context Too close to call
Gemma 2 27B: 37.3 (#218), Mistral Large: 38.3 (#199)
| Benchmark | Gemma 2 27B | Mistral Large |
|---|---|---|
| LMArena Longer Query | 1231 | 1261 |
Writing & Preference Gemma 2 27B leads
Gemma 2 27B: 44.2 (#225), Mistral Large: 40.7 (#242)
| Benchmark | Gemma 2 27B | Mistral Large |
|---|---|---|
| LMArena Text | 1231 | 1266 |
| LMArena Creative Writing | 1241 | 1243 |
| LMArena Multi-Turn | 1224 | 1260 |
| LiveBench Language | 32.6% | 39.4% |
| Short-Story Creative Writing | — | 69% |
| EQ-Bench Creative Writing | — | 985 |
| WildBench | — | 80.1% |
Frequently asked questions
Is Gemma 2 27B better than Mistral Large?
Mistral Large is the stronger model overall, scoring 31.9 to 29.4 on the Noometry Index. Gemma 2 27B costs 4.6× less per token, which makes it the better buy when Mistral Large's lead doesn't matter for your workload.
Which is cheaper, Gemma 2 27B or Mistral Large?
Gemma 2 27B is cheaper. It lists at $0.65 per million input tokens and $0.65 per million output tokens; Mistral Large lists at $2 and $6.
Is Gemma 2 27B or Mistral Large better for coding?
They score almost the same on coding (34.1 vs 34.3); test both on your own repository before choosing.
Which has the bigger context window?
Mistral Large does, with 131K tokens against 8K.
How many benchmarks do Gemma 2 27B and Mistral Large share?
34 benchmarks have published results for both models. Gemma 2 27B has 34 scored results on Noometry and Mistral Large has 51.