Model comparison
Gemma 4 31B IT vs Mistral Large
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 31.9 on the Noometry Index.
Last verified . 24 shared benchmarks.
Google
Mistral AISummary
- They share 24 benchmarks with published results for both. Gemma 4 31B IT scores higher in 8 categories and Mistral Large in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Gemma 4 31B IT leads 43.2 to 18.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 73.3% for Gemma 4 31B IT and 8.5% for Mistral Large.
- Gemma 4 31B IT is cheaper at $0.09 / $0.34 per million input/output tokens, against $2 / $6 for Mistral Large.
- Gemma 4 31B IT accepts more context: 262K tokens versus 131K.
Side by side
| Gemma 4 31B IT | Mistral Large | |
|---|---|---|
| Provider | Mistral AI | |
| Noometry Index | 43.5 | 31.9 |
| Released | 2026-04-02 | 2024-02-26 |
| Weights | Open | Open |
| Context window | 262K | 131K |
| Max output | 33K | 16K |
| Input $ / M tokens | $0.09 | $2 |
| Output $ / M tokens | $0.34 | $6 |
| Results tracked | 35 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 4 31B IT leads
Gemma 4 31B IT: 42.3 (#108), Mistral Large: 34.3 (#240)
| Benchmark | Gemma 4 31B IT | Mistral Large |
|---|---|---|
| SciCode | 43.4% | 36.2% |
| LMArena Coding | 1459 | 1277 |
| ALE-Bench | 925.5 | 264.7 |
| LMArena WebDev | 1366 | — |
| WeirdML | 52.3% | — |
| BigCodeBench Instruct | — | 30% |
| LiveBench Coding | — | 47.1% |
| BigCodeBench Complete | — | 38.3% |
| HumanEval+ | — | 62.2% |
| MBPP+ | — | 59.5% |
Agentic & Tool Use Not comparable
Gemma 4 31B IT: —, Mistral Large: 28.6 (#89)
| Benchmark | Gemma 4 31B IT | Mistral Large |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 38.4% |
Reasoning Gemma 4 31B IT leads
Gemma 4 31B IT: 27.2 (#122), Mistral Large: 15.8 (#310)
| Benchmark | Gemma 4 31B IT | Mistral Large |
|---|---|---|
| CritPt | 1.4% | 0% |
| LMArena Hard Prompts | 1448 | 1257 |
| DTBench | 82.7% | 65.1% |
| LMCA | 39.3% | 16.7% |
| Epoch Capabilities Index | 142.74 | 128.52 |
| SimpleBench | — | 22.5% |
| Kagi LLM Benchmark | 63.5% | — |
| NYT Connections (extended) | 70.6% | — |
| Chess Puzzles | 5% | — |
| Thematic Generalization | 53% | — |
| LiveBench Reasoning | — | 43.5% |
| LiveBench Data Analysis | — | 50.1% |
| Surface Evolver Bench | 30.6% | — |
| ForecastBench | — | 57.1 |
| LiveBench | — | 48.4% |
Math Gemma 4 31B IT leads
Gemma 4 31B IT: 43.2 (#81), Mistral Large: 18.2 (#291)
| Benchmark | Gemma 4 31B IT | Mistral Large |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.3% | 8.5% |
| LMArena Math | 1465 | 1262 |
| Omni-MATH | — | 28.1% |
| LiveBench Math | — | 42.5% |
| MATH Level 5 | — | 50.3% |
| FrontierMath (Feb 2025 set) | — | 0.3% |
Knowledge Gemma 4 31B IT leads
Gemma 4 31B IT: 37.9 (#151), Mistral Large: 30.1 (#230)
| Benchmark | Gemma 4 31B IT | Mistral Large |
|---|---|---|
| GPQA Diamond | 75.8% | 51.3% |
| Vectara Hallucination Rate | 7.4% | 4.5% |
| LMArena Expert | 1465 | 1232 |
| SimpleQA Verified | 10.4% | — |
| MMLU-Pro | — | 59.9% |
| Confabulations | — | 21.4% |
| GPQA (HELM) | — | 43.5% |
| MMLU | — | 80% |
Multimodal Not comparable
Gemma 4 31B IT: 41.6 (#34), Mistral Large: —
| Benchmark | Gemma 4 31B IT | Mistral Large |
|---|---|---|
| LMArena Vision | 1277 | — |
| LMArena Document | 1425 | — |
Multilingual Gemma 4 31B IT leads
Gemma 4 31B IT: 53.8 (#57), Mistral Large: 40.0 (#219)
| Benchmark | Gemma 4 31B IT | Mistral Large |
|---|---|---|
| LMArena Non-English | 1431 | 1237 |
| LMArena Chinese | 1476 | 1240 |
| LMArena French | 1435 | 1325 |
| LMArena Russian | 1460 | 1257 |
| LMArena Spanish | 1444 | 1268 |
| LMArena German | — | 1254 |
| LMArena Japanese | — | 1188 |
| LMArena Korean | — | 1202 |
Instruction Following Gemma 4 31B IT leads
Gemma 4 31B IT: 75.5 (#61), Mistral Large: 67.9 (#191)
| Benchmark | Gemma 4 31B IT | Mistral Large |
|---|---|---|
| LMArena Instruction Following | 1433 | 1249 |
| LiveBench Instruction Following | — | 67.9% |
| IFEval | — | 87.7% |
Long Context Gemma 4 31B IT leads
Gemma 4 31B IT: 44.2 (#71), Mistral Large: 38.3 (#199)
| Benchmark | Gemma 4 31B IT | Mistral Large |
|---|---|---|
| LMArena Longer Query | 1446 | 1261 |
Writing & Preference Gemma 4 31B IT leads
Gemma 4 31B IT: 60.5 (#96), Mistral Large: 40.7 (#242)
| Benchmark | Gemma 4 31B IT | Mistral Large |
|---|---|---|
| LMArena Text | 1443 | 1266 |
| LMArena Creative Writing | 1415 | 1243 |
| EQ-Bench Creative Writing | 1368 | 985 |
| LMArena Multi-Turn | 1452 | 1260 |
| Short-Story Creative Writing | — | 69% |
| WildBench | — | 80.1% |
| EQ-Bench 4 | 1120 | — |
| LiveBench Language | — | 39.4% |
Frequently asked questions
Is Gemma 4 31B IT better than Mistral Large?
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 31.9 on the Noometry Index.
Which is cheaper, Gemma 4 31B IT or Mistral Large?
Gemma 4 31B IT is cheaper. It lists at $0.09 per million input tokens and $0.34 per million output tokens; Mistral Large lists at $2 and $6.
Is Gemma 4 31B IT or Mistral Large better for coding?
Gemma 4 31B IT scores higher on coding benchmarks: 42.3 versus 34.3 in the Noometry coding category.
Which has the bigger context window?
Gemma 4 31B IT does, with 262K tokens against 131K.
How many benchmarks do Gemma 4 31B IT and Mistral Large share?
24 benchmarks have published results for both models. Gemma 4 31B IT has 35 scored results on Noometry and Mistral Large has 51.