Model comparison
Mistral Nemo vs Qwen2.5 72B Instruct
Qwen2.5 72B Instruct is the stronger model overall, scoring 31.9 to 26.4 on the Noometry Index. Mistral Nemo costs 16× less per token, which makes it the better buy when Qwen2.5 72B Instruct's lead doesn't matter for your workload.
Last verified . 6 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 6 benchmarks with published results for both. Mistral Nemo scores higher in 2 categories and Qwen2.5 72B Instruct in 3 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen2.5 72B Instruct leads 46.7 to 28.5.
- The biggest single-benchmark swing is MATH Level 5: 10.8% for Mistral Nemo and 63.2% for Qwen2.5 72B Instruct.
- Mistral Nemo is cheaper at $0.15 / $0.15 per million input/output tokens, against $1.40 / $5.60 for Qwen2.5 72B Instruct.
- Qwen2.5 72B Instruct accepts more context: 131K tokens versus 128K.
Side by side
| Mistral Nemo | Qwen2.5 72B Instruct | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 26.4 | 31.9 |
| Released | 2024-07-01 | 2024-09 |
| Weights | Open | Open |
| Context window | 128K | 131K |
| Max output | 128K | 8K |
| Input $ / M tokens | $0.15 | $1.40 |
| Output $ / M tokens | $0.15 | $5.60 |
| Results tracked | 10 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Mistral Nemo: —, Qwen2.5 72B Instruct: 33.2 (#260)
| Benchmark | Mistral Nemo | Qwen2.5 72B Instruct |
|---|---|---|
| WeirdML | — | 16% |
| BigCodeBench Instruct | — | 45.8% |
| LMArena Coding | — | 1292 |
| BigCodeBench Complete | — | 55.9% |
Agentic & Tool Use Mistral Nemo leads
Mistral Nemo: 23.5 (#125), Qwen2.5 72B Instruct: 22.1 (#133)
| Benchmark | Mistral Nemo | Qwen2.5 72B Instruct |
|---|---|---|
| BALROG | 17.6% | 16.2% |
| Berkeley Function Calling Leaderboard | 27.6% | — |
| TheAgentCompany | — | 5.7% |
| METR Time Horizons | — | 35.8% |
Reasoning Qwen2.5 72B Instruct leads
Mistral Nemo: 20.7 (#232), Qwen2.5 72B Instruct: 22.3 (#199)
| Benchmark | Mistral Nemo | Qwen2.5 72B Instruct |
|---|---|---|
| DTBench | 48.6% | 62.9% |
| Epoch Capabilities Index | 118.68 | 129 |
| PIQA | 83.5% | 82.6% |
| LMArena Hard Prompts | — | 1271 |
| LMCA | — | 13.4% |
| BIG-Bench Hard | — | 79.8% |
| ForecastBench | — | 57.5 |
| HellaSwag | — | 84.8% |
| WinoGrande | — | 82.3% |
Math Mistral Nemo leads
Mistral Nemo: 25.5 (#268), Qwen2.5 72B Instruct: 19.3 (#287)
| Benchmark | Mistral Nemo | Qwen2.5 72B Instruct |
|---|---|---|
| MATH Level 5 | 10.8% | 63.2% |
| OTIS Mock AIME 2024-2025 | — | 8.1% |
| Omni-MATH | — | 33% |
| LMArena Math | — | 1283 |
| GSM8K | 84.2% | — |
Knowledge Qwen2.5 72B Instruct leads
Mistral Nemo: 12.3 (#298), Qwen2.5 72B Instruct: 27.0 (#253)
| Benchmark | Mistral Nemo | Qwen2.5 72B Instruct |
|---|---|---|
| GPQA Diamond | 29.9% | 49.1% |
| MMLU-Pro | — | 63.1% |
| Confabulations | — | 19.1% |
| GPQA (HELM) | — | 42.6% |
| LMArena Expert | — | 1245 |
| ARC (AI2) Challenge | — | 94.5% |
| BoolQ | 82.5% | — |
| MMLU | — | 85.3% |
| TriviaQA | — | 71.9% |
Multilingual Not comparable
Mistral Nemo: —, Qwen2.5 72B Instruct: 41.0 (#213)
| Benchmark | Mistral Nemo | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | — | 1252 |
| LMArena Chinese | — | 1272 |
| LMArena French | — | 1280 |
| LMArena German | — | 1234 |
| LMArena Japanese | — | 1180 |
| LMArena Korean | — | 1188 |
| LMArena Russian | — | 1264 |
| LMArena Spanish | — | 1256 |
Instruction Following Not comparable
Mistral Nemo: —, Qwen2.5 72B Instruct: 65.5 (#221)
| Benchmark | Mistral Nemo | Qwen2.5 72B Instruct |
|---|---|---|
| IFEval | — | 80.6% |
| LMArena Instruction Following | — | 1254 |
Long Context Not comparable
Mistral Nemo: —, Qwen2.5 72B Instruct: 38.9 (#188)
| Benchmark | Mistral Nemo | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | — | 1282 |
Writing & Preference Qwen2.5 72B Instruct leads
Mistral Nemo: 28.5 (#296), Qwen2.5 72B Instruct: 46.7 (#215)
| Benchmark | Mistral Nemo | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | — | 1269 |
| LMArena Creative Writing | — | 1221 |
| EQ-Bench Creative Writing | 881 | — |
| WildBench | — | 80.2% |
| LMArena Multi-Turn | — | 1272 |
Frequently asked questions
Is Mistral Nemo better than Qwen2.5 72B Instruct?
Qwen2.5 72B Instruct is the stronger model overall, scoring 31.9 to 26.4 on the Noometry Index. Mistral Nemo costs 16× less per token, which makes it the better buy when Qwen2.5 72B Instruct's lead doesn't matter for your workload.
Which is cheaper, Mistral Nemo or Qwen2.5 72B Instruct?
Mistral Nemo is cheaper. It lists at $0.15 per million input tokens and $0.15 per million output tokens; Qwen2.5 72B Instruct lists at $1.40 and $5.60.
Which has the bigger context window?
Qwen2.5 72B Instruct does, with 131K tokens against 128K.
How many benchmarks do Mistral Nemo and Qwen2.5 72B Instruct share?
6 benchmarks have published results for both models. Mistral Nemo has 10 scored results on Noometry and Qwen2.5 72B Instruct has 43.