Model comparison
Codestral vs Qwen2.5 72B Instruct
Qwen2.5 72B Instruct is the stronger model overall, scoring 31.9 to 30.6 on the Noometry Index. Codestral costs 5.4× less per token, which makes it the better buy when Qwen2.5 72B Instruct's lead doesn't matter for your workload.
Last verified . 2 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 2 benchmarks with published results for both. Codestral scores higher in 0 categories and Qwen2.5 72B Instruct in 2 categories; 2 gaps are clear of the uncertainty.
- The widest gap is in coding, where Qwen2.5 72B Instruct leads 33.2 to 27.3.
- Codestral is cheaper at $0.30 / $0.90 per million input/output tokens, against $1.40 / $5.60 for Qwen2.5 72B Instruct.
- Codestral accepts more context: 256K tokens versus 131K.
- Qwen2.5 72B Instruct has downloadable open weights; the other is API-only.
Side by side
| Codestral | Qwen2.5 72B Instruct | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 30.6 | 31.9 |
| Released | 2024-05-29 | 2024-09 |
| Weights | Proprietary | Open |
| Context window | 256K | 131K |
| Max output | 8K | 8K |
| Input $ / M tokens | $0.30 | $1.40 |
| Output $ / M tokens | $0.90 | $5.60 |
| Results tracked | 7 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen2.5 72B Instruct leads
Codestral: 27.3 (#321), Qwen2.5 72B Instruct: 33.2 (#260)
| Benchmark | Codestral | Qwen2.5 72B Instruct |
|---|---|---|
| BigCodeBench Instruct | 41.8% | 45.8% |
| BigCodeBench Complete | 52.5% | 55.9% |
| Aider Polyglot | 11.1% | — |
| WeirdML | — | 16% |
| LMArena Coding | — | 1292 |
| ALE-Bench | 137.78 | — |
| HumanEval+ | 73.8% | — |
| MBPP+ | 61.9% | — |
Agentic & Tool Use Not comparable
Codestral: —, Qwen2.5 72B Instruct: 22.1 (#133)
| Benchmark | Codestral | Qwen2.5 72B Instruct |
|---|---|---|
| TheAgentCompany | — | 5.7% |
| BALROG | — | 16.2% |
| METR Time Horizons | — | 35.8% |
Reasoning Qwen2.5 72B Instruct leads
Codestral: 19.8 (#251), Qwen2.5 72B Instruct: 22.3 (#199)
| Benchmark | Codestral | Qwen2.5 72B Instruct |
|---|---|---|
| Kagi LLM Benchmark | 32.5% | — |
| LMArena Hard Prompts | — | 1271 |
| DTBench | — | 62.9% |
| LMCA | — | 13.4% |
| BIG-Bench Hard | — | 79.8% |
| Epoch Capabilities Index | — | 129 |
| ForecastBench | — | 57.5 |
| HellaSwag | — | 84.8% |
| PIQA | — | 82.6% |
| WinoGrande | — | 82.3% |
Math Not comparable
Codestral: —, Qwen2.5 72B Instruct: 19.3 (#287)
| Benchmark | Codestral | Qwen2.5 72B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 8.1% |
| Omni-MATH | — | 33% |
| LMArena Math | — | 1283 |
| MATH Level 5 | — | 63.2% |
Knowledge Not comparable
Codestral: —, Qwen2.5 72B Instruct: 27.0 (#253)
| Benchmark | Codestral | Qwen2.5 72B Instruct |
|---|---|---|
| GPQA Diamond | — | 49.1% |
| MMLU-Pro | — | 63.1% |
| Confabulations | — | 19.1% |
| GPQA (HELM) | — | 42.6% |
| LMArena Expert | — | 1245 |
| ARC (AI2) Challenge | — | 94.5% |
| MMLU | — | 85.3% |
| TriviaQA | — | 71.9% |
Multilingual Not comparable
Codestral: —, Qwen2.5 72B Instruct: 41.0 (#213)
| Benchmark | Codestral | 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
Codestral: —, Qwen2.5 72B Instruct: 65.5 (#221)
| Benchmark | Codestral | Qwen2.5 72B Instruct |
|---|---|---|
| IFEval | — | 80.6% |
| LMArena Instruction Following | — | 1254 |
Long Context Not comparable
Codestral: —, Qwen2.5 72B Instruct: 38.9 (#188)
| Benchmark | Codestral | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | — | 1282 |
Writing & Preference Not comparable
Codestral: —, Qwen2.5 72B Instruct: 46.7 (#215)
| Benchmark | Codestral | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | — | 1269 |
| LMArena Creative Writing | — | 1221 |
| WildBench | — | 80.2% |
| LMArena Multi-Turn | — | 1272 |
Frequently asked questions
Is Codestral better than Qwen2.5 72B Instruct?
Qwen2.5 72B Instruct is the stronger model overall, scoring 31.9 to 30.6 on the Noometry Index. Codestral costs 5.4× 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, Codestral or Qwen2.5 72B Instruct?
Codestral is cheaper. It lists at $0.30 per million input tokens and $0.90 per million output tokens; Qwen2.5 72B Instruct lists at $1.40 and $5.60.
Is Codestral or Qwen2.5 72B Instruct better for coding?
Qwen2.5 72B Instruct scores higher on coding benchmarks: 33.2 versus 27.3 in the Noometry coding category.
Which has the bigger context window?
Codestral does, with 256K tokens against 131K.
How many benchmarks do Codestral and Qwen2.5 72B Instruct share?
2 benchmarks have published results for both models. Codestral has 7 scored results on Noometry and Qwen2.5 72B Instruct has 43.