Model comparison
Mistral Small vs Qwen2.5 7B Instruct
Mistral Small is the stronger model overall, scoring 33.4 to 29.0 on the Noometry Index.
Last verified . 7 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 7 benchmarks with published results for both. Mistral Small scores higher in 6 categories and Qwen2.5 7B Instruct in 1 category; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Mistral Small leads 31.0 to 17.0.
- The biggest single-benchmark swing is DTBench: 70.9% for Mistral Small and 47.7% for Qwen2.5 7B Instruct.
- Mistral Small is cheaper at $0.15 / $0.60 per million input/output tokens, against $0.17 / $0.70 for Qwen2.5 7B Instruct.
- Mistral Small accepts more context: 262K tokens versus 131K.
Side by side
| Mistral Small | Qwen2.5 7B Instruct | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 33.4 | 29.0 |
| Released | 2024-02-26 | 2024-09 |
| Weights | Open | Open |
| Context window | 262K | 131K |
| Max output | 256K | 8K |
| Input $ / M tokens | $0.15 | $0.17 |
| Output $ / M tokens | $0.60 | $0.70 |
| Results tracked | 39 | 15 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen2.5 7B Instruct leads
Mistral Small: 34.0 (#247), Qwen2.5 7B Instruct: 36.5 (#208)
| Benchmark | Mistral Small | Qwen2.5 7B Instruct |
|---|---|---|
| BigCodeBench Instruct | 36.1% | 37.6% |
| BigCodeBench Complete | 46.6% | 46.1% |
| SciCode | 26.5% | — |
| LiveBench Coding | 36.2% | — |
| LMArena Coding | 1362 | — |
| ALE-Bench | 497.62 | — |
Agentic & Tool Use Mistral Small leads
Mistral Small: 28.1 (#93), Qwen2.5 7B Instruct: 23.8 (#124)
| Benchmark | Mistral Small | Qwen2.5 7B Instruct |
|---|---|---|
| Berkeley Function Calling Leaderboard | 37.1% | — |
| BALROG | — | 7.8% |
Reasoning Mistral Small leads
Mistral Small: 19.8 (#250), Qwen2.5 7B Instruct: 14.8 (#322)
| Benchmark | Mistral Small | Qwen2.5 7B Instruct |
|---|---|---|
| DTBench | 70.9% | 47.7% |
| LMCA | 20.6% | 6.4% |
| Kagi LLM Benchmark | 37.8% | — |
| CritPt | 0% | — |
| Chess Puzzles | — | 0% |
| LiveBench Reasoning | 44.8% | — |
| LMArena Hard Prompts | 1335 | — |
| LiveBench Data Analysis | 53.7% | — |
| Epoch Capabilities Index | — | 118.51 |
| LiveBench | 44% | — |
Math Mistral Small leads
Mistral Small: 16.4 (#293), Qwen2.5 7B Instruct: 12.6 (#306)
| Benchmark | Mistral Small | Qwen2.5 7B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 5.8% | 2.5% |
| Omni-MATH | — | 29.4% |
| LiveBench Math | 39.9% | — |
| LMArena Math | 1341 | — |
| MATH Level 5 | 46.8% | — |
Knowledge Mistral Small leads
Mistral Small: 31.0 (#222), Qwen2.5 7B Instruct: 17.0 (#286)
| Benchmark | Mistral Small | Qwen2.5 7B Instruct |
|---|---|---|
| GPQA Diamond | 47.5% | 35.5% |
| MMLU | 68.7% | 72.9% |
| MMLU-Pro | — | 53.9% |
| Vectara Hallucination Rate | 5.1% | — |
| GPQA (HELM) | — | 34.1% |
| LMArena Expert | 1291 | — |
Multimodal Not comparable
Mistral Small: 33.5 (#96), Qwen2.5 7B Instruct: —
| Benchmark | Mistral Small | Qwen2.5 7B Instruct |
|---|---|---|
| LMArena Vision | 1142 | — |
Multilingual Not comparable
Mistral Small: 45.5 (#169), Qwen2.5 7B Instruct: —
| Benchmark | Mistral Small | Qwen2.5 7B Instruct |
|---|---|---|
| LMArena Non-English | 1315 | — |
| LMArena Chinese | 1340 | — |
| LMArena French | 1337 | — |
| LMArena German | 1340 | — |
| LMArena Japanese | 1275 | — |
| LMArena Korean | 1259 | — |
| LMArena Russian | 1324 | — |
| LMArena Spanish | 1346 | — |
Instruction Following Mistral Small leads
Mistral Small: 66.4 (#209), Qwen2.5 7B Instruct: 63.2 (#231)
| Benchmark | Mistral Small | Qwen2.5 7B Instruct |
|---|---|---|
| LiveBench Instruction Following | 63.7% | — |
| IFEval | — | 74.1% |
| LMArena Instruction Following | 1310 | — |
Long Context Not comparable
Mistral Small: 40.4 (#156), Qwen2.5 7B Instruct: —
| Benchmark | Mistral Small | Qwen2.5 7B Instruct |
|---|---|---|
| LMArena Longer Query | 1327 | — |
Writing & Preference Mistral Small leads
Mistral Small: 52.5 (#171), Qwen2.5 7B Instruct: 48.8 (#195)
| Benchmark | Mistral Small | Qwen2.5 7B Instruct |
|---|---|---|
| LMArena Text | 1338 | — |
| LMArena Creative Writing | 1305 | — |
| WildBench | — | 73.1% |
| LMArena Multi-Turn | 1344 | — |
| LiveBench Language | 30.5% | — |
Frequently asked questions
Is Mistral Small better than Qwen2.5 7B Instruct?
Mistral Small is the stronger model overall, scoring 33.4 to 29.0 on the Noometry Index.
Which is cheaper, Mistral Small or Qwen2.5 7B Instruct?
Mistral Small is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Qwen2.5 7B Instruct lists at $0.17 and $0.70.
Is Mistral Small or Qwen2.5 7B Instruct better for coding?
Qwen2.5 7B Instruct scores higher on coding benchmarks: 36.5 versus 34.0 in the Noometry coding category.
Which has the bigger context window?
Mistral Small does, with 262K tokens against 131K.
How many benchmarks do Mistral Small and Qwen2.5 7B Instruct share?
7 benchmarks have published results for both models. Mistral Small has 39 scored results on Noometry and Qwen2.5 7B Instruct has 15.