Model comparison
Mistral Large vs Qwen2-72B
Mistral Large is the stronger model overall, scoring 31.9 to 30.0 on the Noometry Index.
Last verified . 23 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 23 benchmarks with published results for both. Mistral Large scores higher in 6 categories and Qwen2-72B in 3 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen2-72B leads 30.2 to 18.2.
- The biggest single-benchmark swing is BigCodeBench Complete: 38.3% for Mistral Large and 54% for Qwen2-72B.
Side by side
| Mistral Large | Qwen2-72B | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 31.9 | 30.0 |
| Released | 2024-02-26 | 2024-06-07 |
| Weights | Open | Open |
| Context window | 131K | — |
| Max output | 16K | — |
| Input $ / M tokens | $2 | — |
| Output $ / M tokens | $6 | — |
| Results tracked | 51 | 26 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mistral Large leads
Mistral Large: 34.3 (#240), Qwen2-72B: 29.1 (#310)
| Benchmark | Mistral Large | Qwen2-72B |
|---|---|---|
| BigCodeBench Instruct | 30% | 38.5% |
| LMArena Coding | 1277 | 1196 |
| BigCodeBench Complete | 38.3% | 54% |
| SciCode | 36.2% | — |
| WeirdML | — | 11.3% |
| LiveBench Coding | 47.1% | — |
| ALE-Bench | 264.7 | — |
| HumanEval+ | 62.2% | — |
| MBPP+ | 59.5% | — |
Agentic & Tool Use Mistral Large leads
Mistral Large: 28.6 (#89), Qwen2-72B: 17.0 (#146)
| Benchmark | Mistral Large | Qwen2-72B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 38.4% | — |
| TheAgentCompany | — | 1.1% |
| METR Time Horizons | — | 29.9% |
Reasoning Qwen2-72B leads
Mistral Large: 15.8 (#310), Qwen2-72B: 23.2 (#181)
| Benchmark | Mistral Large | Qwen2-72B |
|---|---|---|
| LMArena Hard Prompts | 1257 | 1191 |
| Epoch Capabilities Index | 128.52 | 125.28 |
| SimpleBench | 22.5% | — |
| CritPt | 0% | — |
| LiveBench Reasoning | 43.5% | — |
| DTBench | 65.1% | — |
| LiveBench Data Analysis | 50.1% | — |
| LMCA | 16.7% | — |
| ForecastBench | 57.1 | — |
| LiveBench | 48.4% | — |
Math Qwen2-72B leads
Mistral Large: 18.2 (#291), Qwen2-72B: 30.2 (#236)
| Benchmark | Mistral Large | Qwen2-72B |
|---|---|---|
| LMArena Math | 1262 | 1235 |
| MATH Level 5 | 50.3% | 39.1% |
| OTIS Mock AIME 2024-2025 | 8.5% | — |
| Omni-MATH | 28.1% | — |
| LiveBench Math | 42.5% | — |
| FrontierMath (Feb 2025 set) | 0.3% | — |
Knowledge Mistral Large leads
Mistral Large: 30.1 (#230), Qwen2-72B: 21.2 (#275)
| Benchmark | Mistral Large | Qwen2-72B |
|---|---|---|
| GPQA Diamond | 51.3% | 40.8% |
| LMArena Expert | 1232 | 1171 |
| MMLU | 80% | 82.4% |
| MMLU-Pro | 59.9% | — |
| Confabulations | 21.4% | — |
| Vectara Hallucination Rate | 4.5% | — |
| GPQA (HELM) | 43.5% | — |
Multilingual Mistral Large leads
Mistral Large: 40.0 (#219), Qwen2-72B: 35.9 (#244)
| Benchmark | Mistral Large | Qwen2-72B |
|---|---|---|
| LMArena Non-English | 1237 | 1176 |
| LMArena Chinese | 1240 | 1240 |
| LMArena French | 1325 | 1170 |
| LMArena German | 1254 | 1151 |
| LMArena Japanese | 1188 | 1111 |
| LMArena Korean | 1202 | 1083 |
| LMArena Russian | 1257 | 1169 |
| LMArena Spanish | 1268 | 1169 |
Instruction Following Mistral Large leads
Mistral Large: 67.9 (#191), Qwen2-72B: 61.7 (#241)
| Benchmark | Mistral Large | Qwen2-72B |
|---|---|---|
| LMArena Instruction Following | 1249 | 1181 |
| LiveBench Instruction Following | 67.9% | — |
| IFEval | 87.7% | — |
Long Context Mistral Large leads
Mistral Large: 38.3 (#199), Qwen2-72B: 36.1 (#235)
| Benchmark | Mistral Large | Qwen2-72B |
|---|---|---|
| LMArena Longer Query | 1261 | 1192 |
Writing & Preference Too close to call
Mistral Large: 40.7 (#242), Qwen2-72B: 40.8 (#241)
| Benchmark | Mistral Large | Qwen2-72B |
|---|---|---|
| LMArena Text | 1266 | 1203 |
| LMArena Creative Writing | 1243 | 1181 |
| LMArena Multi-Turn | 1260 | 1196 |
| Short-Story Creative Writing | 69% | — |
| EQ-Bench Creative Writing | 985 | — |
| WildBench | 80.1% | — |
| LiveBench Language | 39.4% | — |
Frequently asked questions
Is Mistral Large better than Qwen2-72B?
Mistral Large is the stronger model overall, scoring 31.9 to 30.0 on the Noometry Index.
Is Mistral Large or Qwen2-72B better for coding?
Mistral Large scores higher on coding benchmarks: 34.3 versus 29.1 in the Noometry coding category.
How many benchmarks do Mistral Large and Qwen2-72B share?
23 benchmarks have published results for both models. Mistral Large has 51 scored results on Noometry and Qwen2-72B has 26.