Model comparison
Mistral Small 3 vs Qwen2-72B
Mistral Small 3 is the stronger model overall, scoring 30.9 to 29.5 on the Noometry Index.
Last verified . 20 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 20 benchmarks with published results for both. Mistral Small 3 scores higher in 5 categories and Qwen2-72B in 3 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen2-72B leads 27.9 to 15.3.
- The biggest single-benchmark swing is BigCodeBench Instruct: 45.3% for Mistral Small 3 and 38.5% for Qwen2-72B.
Side by side
| Mistral Small 3 | Qwen2-72B | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 30.9 | 29.5 |
| Released | 2025-01-30 | 2024-06-07 |
| Weights | Open | Open |
| Context window | 33K | — |
| Max output | 16K | — |
| Input $ / M tokens | $0.05 | — |
| Output $ / M tokens | $0.08 | — |
| Results tracked | 24 | 26 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mistral Small 3 leads
Mistral Small 3: 36.0 (#203), Qwen2-72B: 28.7 (#303)
| Benchmark | Mistral Small 3 | Qwen2-72B |
|---|---|---|
| BigCodeBench Instruct | 45.3% | 38.5% |
| LMArena Coding | 1246 | 1196 |
| BigCodeBench Complete | 50.4% | 54% |
| WeirdML | — | 11.3% |
Agentic & Tool Use Not comparable
Mistral Small 3: —, Qwen2-72B: 16.9 (#138)
| Benchmark | Mistral Small 3 | Qwen2-72B |
|---|---|---|
| TheAgentCompany | — | 1.1% |
| METR Time Horizons | — | 29.9% |
Reasoning Qwen2-72B leads
Mistral Small 3: 18.2 (#269), Qwen2-72B: 22.2 (#184)
| Benchmark | Mistral Small 3 | Qwen2-72B |
|---|---|---|
| LMArena Hard Prompts | 1233 | 1191 |
| Epoch Capabilities Index | 127.06 | 125.27 |
| Chess Puzzles | 0% | — |
Math Qwen2-72B leads
Mistral Small 3: 15.3 (#290), Qwen2-72B: 27.9 (#232)
| Benchmark | Mistral Small 3 | Qwen2-72B |
|---|---|---|
| LMArena Math | 1240 | 1235 |
| OTIS Mock AIME 2024-2025 | 6.7% | — |
| MATH Level 5 | — | 39.1% |
Knowledge Mistral Small 3 leads
Mistral Small 3: 25.0 (#258), Qwen2-72B: 21.2 (#270)
| Benchmark | Mistral Small 3 | Qwen2-72B |
|---|---|---|
| GPQA Diamond | 47.3% | 40.8% |
| LMArena Expert | 1202 | 1171 |
| Confabulations | 25.2% | — |
| MMLU | — | 82.4% |
Multilingual Mistral Small 3 leads
Mistral Small 3: 37.5 (#231), Qwen2-72B: 36.0 (#239)
| Benchmark | Mistral Small 3 | Qwen2-72B |
|---|---|---|
| LMArena Non-English | 1198 | 1176 |
| LMArena Chinese | 1204 | 1240 |
| LMArena French | 1202 | 1169 |
| LMArena German | 1211 | 1151 |
| LMArena Japanese | 1111 | 1111 |
| LMArena Korean | 1188 | 1083 |
| LMArena Russian | 1216 | 1169 |
| LMArena Spanish | — | 1168 |
Instruction Following Mistral Small 3 leads
Mistral Small 3: 63.9 (#224), Qwen2-72B: 62.0 (#236)
| Benchmark | Mistral Small 3 | Qwen2-72B |
|---|---|---|
| LMArena Instruction Following | 1214 | 1181 |
Long Context Mistral Small 3 leads
Mistral Small 3: 37.8 (#206), Qwen2-72B: 36.1 (#230)
| Benchmark | Mistral Small 3 | Qwen2-72B |
|---|---|---|
| LMArena Longer Query | 1246 | 1193 |
Writing & Preference Qwen2-72B leads
Mistral Small 3: 32.4 (#275), Qwen2-72B: 41.0 (#236)
| Benchmark | Mistral Small 3 | Qwen2-72B |
|---|---|---|
| LMArena Text | 1234 | 1203 |
| LMArena Creative Writing | 1195 | 1181 |
| LMArena Multi-Turn | 1217 | 1195 |
| EQ-Bench Creative Writing | 707 | — |
Frequently asked questions
Is Mistral Small 3 better than Qwen2-72B?
Mistral Small 3 is the stronger model overall, scoring 30.9 to 29.5 on the Noometry Index.
Is Mistral Small 3 or Qwen2-72B better for coding?
Mistral Small 3 scores higher on coding benchmarks: 36.0 versus 28.7 in the Noometry coding category.
How many benchmarks do Mistral Small 3 and Qwen2-72B share?
20 benchmarks have published results for both models. Mistral Small 3 has 24 scored results on Noometry and Qwen2-72B has 26.