Model comparison
Codestral vs Qwen Max
Qwen Max is the stronger model overall, scoring 34.7 to 30.6 on the Noometry Index. Codestral costs 6.2× less per token, which makes it the better buy when Qwen Max's lead doesn't matter for your workload.
Last verified . 1 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 1 benchmark with published results for both. Codestral scores higher in 0 categories and Qwen Max in 2 categories; 2 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen Max leads 25.1 to 19.8.
- The biggest single-benchmark swing is Aider Polyglot: 11.1% for Codestral and 21.8% for Qwen Max.
- Codestral is cheaper at $0.30 / $0.90 per million input/output tokens, against $1.60 / $6.40 for Qwen Max.
- Codestral accepts more context: 256K tokens versus 33K.
Side by side
| Codestral | Qwen Max | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 30.6 | 34.7 |
| Released | 2024-05-29 | 2024-04-03 |
| Weights | Proprietary | Proprietary |
| Context window | 256K | 33K |
| Max output | 8K | 8K |
| Input $ / M tokens | $0.30 | $1.60 |
| Output $ / M tokens | $0.90 | $6.40 |
| Results tracked | 7 | 23 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen Max leads
Codestral: 27.3 (#321), Qwen Max: 30.7 (#292)
| Benchmark | Codestral | Qwen Max |
|---|---|---|
| Aider Polyglot | 11.1% | 21.8% |
| BigCodeBench Instruct | 41.8% | — |
| LMArena Coding | — | 1288 |
| BigCodeBench Complete | 52.5% | — |
| ALE-Bench | 137.78 | — |
| HumanEval+ | 73.8% | — |
| MBPP+ | 61.9% | — |
Reasoning Qwen Max leads
Codestral: 19.8 (#251), Qwen Max: 25.1 (#151)
| Benchmark | Codestral | Qwen Max |
|---|---|---|
| Kagi LLM Benchmark | 32.5% | — |
| LMArena Hard Prompts | — | 1269 |
Math Not comparable
Codestral: —, Qwen Max: 22.3 (#276)
| Benchmark | Codestral | Qwen Max |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 16.1% |
| LMArena Math | — | 1275 |
| MATH Level 5 | — | 67.2% |
| FrontierMath (Feb 2025 set) | — | 1% |
Knowledge Not comparable
Codestral: —, Qwen Max: 30.3 (#228)
| Benchmark | Codestral | Qwen Max |
|---|---|---|
| GPQA Diamond | — | 56.1% |
| LMArena Expert | — | 1248 |
Multilingual Not comparable
Codestral: —, Qwen Max: 41.8 (#202)
| Benchmark | Codestral | Qwen Max |
|---|---|---|
| LMArena Non-English | — | 1263 |
| LMArena Chinese | — | 1254 |
| LMArena French | — | 1330 |
| LMArena German | — | 1254 |
| LMArena Japanese | — | 1205 |
| LMArena Korean | — | 1142 |
| LMArena Russian | — | 1274 |
| LMArena Spanish | — | 1290 |
Instruction Following Not comparable
Codestral: —, Qwen Max: 66.5 (#208)
| Benchmark | Codestral | Qwen Max |
|---|---|---|
| LMArena Instruction Following | — | 1262 |
Long Context Not comparable
Codestral: —, Qwen Max: 39.4 (#180)
| Benchmark | Codestral | Qwen Max |
|---|---|---|
| Fiction.LiveBench | — | 66.7% |
| LMArena Longer Query | — | 1288 |
Writing & Preference Not comparable
Codestral: —, Qwen Max: 47.8 (#205)
| Benchmark | Codestral | Qwen Max |
|---|---|---|
| LMArena Text | — | 1282 |
| LMArena Creative Writing | — | 1248 |
| LMArena Multi-Turn | — | 1277 |
Frequently asked questions
Is Codestral better than Qwen Max?
Qwen Max is the stronger model overall, scoring 34.7 to 30.6 on the Noometry Index. Codestral costs 6.2× less per token, which makes it the better buy when Qwen Max's lead doesn't matter for your workload.
Which is cheaper, Codestral or Qwen Max?
Codestral is cheaper. It lists at $0.30 per million input tokens and $0.90 per million output tokens; Qwen Max lists at $1.60 and $6.40.
Is Codestral or Qwen Max better for coding?
Qwen Max scores higher on coding benchmarks: 30.7 versus 27.3 in the Noometry coding category.
Which has the bigger context window?
Codestral does, with 256K tokens against 33K.
How many benchmarks do Codestral and Qwen Max share?
1 benchmark has published results for both models. Codestral has 7 scored results on Noometry and Qwen Max has 23.