Model comparison
Mistral Large vs Qwen2.5 7B Instruct
Mistral Large is the stronger model overall, scoring 31.9 to 29.0 on the Noometry Index. Qwen2.5 7B Instruct costs 9.8× less per token, which makes it the better buy when Mistral Large's lead doesn't matter for your workload.
Last verified . 13 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 13 benchmarks with published results for both. Mistral Large scores higher in 5 categories and Qwen2.5 7B Instruct in 2 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Mistral Large leads 30.1 to 17.0.
- The biggest single-benchmark swing is DTBench: 65.1% for Mistral Large and 47.7% for Qwen2.5 7B Instruct.
- Qwen2.5 7B Instruct is cheaper at $0.17 / $0.70 per million input/output tokens, against $2 / $6 for Mistral Large.
Side by side
| Mistral Large | Qwen2.5 7B Instruct | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 31.9 | 29.0 |
| Released | 2024-02-26 | 2024-09 |
| Weights | Open | Open |
| Context window | 131K | 131K |
| Max output | 16K | 8K |
| Input $ / M tokens | $2 | $0.17 |
| Output $ / M tokens | $6 | $0.70 |
| Results tracked | 51 | 15 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen2.5 7B Instruct leads
Mistral Large: 34.3 (#240), Qwen2.5 7B Instruct: 36.5 (#208)
| Benchmark | Mistral Large | Qwen2.5 7B Instruct |
|---|---|---|
| BigCodeBench Instruct | 30% | 37.6% |
| BigCodeBench Complete | 38.3% | 46.1% |
| SciCode | 36.2% | — |
| LiveBench Coding | 47.1% | — |
| LMArena Coding | 1277 | — |
| ALE-Bench | 264.7 | — |
| HumanEval+ | 62.2% | — |
| MBPP+ | 59.5% | — |
Agentic & Tool Use Mistral Large leads
Mistral Large: 28.6 (#89), Qwen2.5 7B Instruct: 23.8 (#124)
| Benchmark | Mistral Large | Qwen2.5 7B Instruct |
|---|---|---|
| Berkeley Function Calling Leaderboard | 38.4% | — |
| BALROG | — | 7.8% |
Reasoning Too close to call
Mistral Large: 15.8 (#310), Qwen2.5 7B Instruct: 14.8 (#322)
| Benchmark | Mistral Large | Qwen2.5 7B Instruct |
|---|---|---|
| DTBench | 65.1% | 47.7% |
| LMCA | 16.7% | 6.4% |
| Epoch Capabilities Index | 128.52 | 118.51 |
| SimpleBench | 22.5% | — |
| CritPt | 0% | — |
| Chess Puzzles | — | 0% |
| LiveBench Reasoning | 43.5% | — |
| LMArena Hard Prompts | 1257 | — |
| LiveBench Data Analysis | 50.1% | — |
| ForecastBench | 57.1 | — |
| LiveBench | 48.4% | — |
Math Mistral Large leads
Mistral Large: 18.2 (#291), Qwen2.5 7B Instruct: 12.6 (#306)
| Benchmark | Mistral Large | Qwen2.5 7B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 8.5% | 2.5% |
| Omni-MATH | 28.1% | 29.4% |
| LiveBench Math | 42.5% | — |
| LMArena Math | 1262 | — |
| MATH Level 5 | 50.3% | — |
| FrontierMath (Feb 2025 set) | 0.3% | — |
Knowledge Mistral Large leads
Mistral Large: 30.1 (#230), Qwen2.5 7B Instruct: 17.0 (#286)
| Benchmark | Mistral Large | Qwen2.5 7B Instruct |
|---|---|---|
| GPQA Diamond | 51.3% | 35.5% |
| MMLU-Pro | 59.9% | 53.9% |
| GPQA (HELM) | 43.5% | 34.1% |
| MMLU | 80% | 72.9% |
| Confabulations | 21.4% | — |
| Vectara Hallucination Rate | 4.5% | — |
| LMArena Expert | 1232 | — |
Multilingual Not comparable
Mistral Large: 40.0 (#219), Qwen2.5 7B Instruct: —
| Benchmark | Mistral Large | Qwen2.5 7B Instruct |
|---|---|---|
| LMArena Non-English | 1237 | — |
| LMArena Chinese | 1240 | — |
| LMArena French | 1325 | — |
| LMArena German | 1254 | — |
| LMArena Japanese | 1188 | — |
| LMArena Korean | 1202 | — |
| LMArena Russian | 1257 | — |
| LMArena Spanish | 1268 | — |
Instruction Following Mistral Large leads
Mistral Large: 67.9 (#191), Qwen2.5 7B Instruct: 63.2 (#231)
| Benchmark | Mistral Large | Qwen2.5 7B Instruct |
|---|---|---|
| IFEval | 87.7% | 74.1% |
| LiveBench Instruction Following | 67.9% | — |
| LMArena Instruction Following | 1249 | — |
Long Context Not comparable
Mistral Large: 38.3 (#199), Qwen2.5 7B Instruct: —
| Benchmark | Mistral Large | Qwen2.5 7B Instruct |
|---|---|---|
| LMArena Longer Query | 1261 | — |
Writing & Preference Qwen2.5 7B Instruct leads
Mistral Large: 40.7 (#242), Qwen2.5 7B Instruct: 48.8 (#195)
| Benchmark | Mistral Large | Qwen2.5 7B Instruct |
|---|---|---|
| WildBench | 80.1% | 73.1% |
| LMArena Text | 1266 | — |
| LMArena Creative Writing | 1243 | — |
| Short-Story Creative Writing | 69% | — |
| EQ-Bench Creative Writing | 985 | — |
| LMArena Multi-Turn | 1260 | — |
| LiveBench Language | 39.4% | — |
Frequently asked questions
Is Mistral Large better than Qwen2.5 7B Instruct?
Mistral Large is the stronger model overall, scoring 31.9 to 29.0 on the Noometry Index. Qwen2.5 7B Instruct costs 9.8× less per token, which makes it the better buy when Mistral Large's lead doesn't matter for your workload.
Which is cheaper, Mistral Large or Qwen2.5 7B Instruct?
Qwen2.5 7B Instruct is cheaper. It lists at $0.17 per million input tokens and $0.70 per million output tokens; Mistral Large lists at $2 and $6.
Is Mistral Large or Qwen2.5 7B Instruct better for coding?
Qwen2.5 7B Instruct scores higher on coding benchmarks: 36.5 versus 34.3 in the Noometry coding category.
Which has the bigger context window?
Both accept 131K tokens.
How many benchmarks do Mistral Large and Qwen2.5 7B Instruct share?
13 benchmarks have published results for both models. Mistral Large has 51 scored results on Noometry and Qwen2.5 7B Instruct has 15.