Model comparison
Codestral vs Qwen3.8 27B
Qwen3.8 27B is the stronger model overall, scoring 45.3 to 30.1 on the Noometry Index. Codestral costs 2.5× less per token, which makes it the better buy when Qwen3.8 27B's lead doesn't matter for your workload.
Last verified . 0 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- The widest gap is in coding, where Qwen3.8 27B leads 50.3 to 26.9.
- Codestral is cheaper at $0.30 / $0.90 per million input/output tokens, against $0.99 / $1.49 for Qwen3.8 27B.
- Qwen3.8 27B accepts more context: 262K tokens versus 256K.
Side by side
| Codestral | Qwen3.8 27B | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 30.1 | 45.3 |
| Released | 2024-05-29 | 2026-08-14 |
| Weights | Open | Open |
| Context window | 256K | 262K |
| Max output | 4K | 33K |
| Input $ / M tokens | $0.30 | $0.99 |
| Output $ / M tokens | $0.90 | $1.49 |
| Results tracked | 7 | 28 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 27B leads
Codestral: 26.9 (#316), Qwen3.8 27B: 50.3 (#42)
| Benchmark | Codestral | Qwen3.8 27B |
|---|---|---|
| Aider Polyglot | 11.1% | — |
| LMArena WebDev | — | 1592 |
| SciCode | — | 46.6% |
| BigCodeBench Instruct | 41.8% | — |
| LMArena Coding | — | 1487 |
| BigCodeBench Complete | 52.5% | — |
| ALE-Bench | 137.78 | — |
| HumanEval+ | 73.8% | — |
| MBPP+ | 61.9% | — |
Reasoning Qwen3.8 27B leads
Codestral: 19.1 (#244), Qwen3.8 27B: 31.5 (#71)
| Benchmark | Codestral | Qwen3.8 27B |
|---|---|---|
| Kagi LLM Benchmark | 32.5% | — |
| NYT Connections (extended) | — | 54.5% |
| CritPt | — | 5.4% |
| LMArena Hard Prompts | — | 1459 |
| DTBench | — | 88% |
| LMCA | — | 41.4% |
| Surface Evolver Bench | — | 45% |
| Epoch Capabilities Index | — | 149.38 |
Math Not comparable
Codestral: —, Qwen3.8 27B: 35.7 (#139)
| Benchmark | Codestral | Qwen3.8 27B |
|---|---|---|
| ProofBench | — | 16% |
| LMArena Math | — | 1453 |
Knowledge Not comparable
Codestral: —, Qwen3.8 27B: 40.9 (#105)
| Benchmark | Codestral | Qwen3.8 27B |
|---|---|---|
| LMArena Expert | — | 1482 |
Multimodal Not comparable
Codestral: —, Qwen3.8 27B: 40.5 (#36)
| Benchmark | Codestral | Qwen3.8 27B |
|---|---|---|
| LMArena Vision | — | 1270 |
Multilingual Not comparable
Codestral: —, Qwen3.8 27B: 54.0 (#54)
| Benchmark | Codestral | Qwen3.8 27B |
|---|---|---|
| LMArena Non-English | — | 1431 |
| LMArena Chinese | — | 1504 |
| LMArena French | — | 1468 |
| LMArena German | — | 1435 |
| LMArena Japanese | — | 1392 |
| LMArena Korean | — | 1389 |
| LMArena Russian | — | 1421 |
| LMArena Spanish | — | 1444 |
Instruction Following Not comparable
Codestral: —, Qwen3.8 27B: 75.8 (#54)
| Benchmark | Codestral | Qwen3.8 27B |
|---|---|---|
| LMArena Instruction Following | — | 1436 |
Long Context Not comparable
Codestral: —, Qwen3.8 27B: 44.3 (#66)
| Benchmark | Codestral | Qwen3.8 27B |
|---|---|---|
| LMArena Longer Query | — | 1448 |
Writing & Preference Not comparable
Codestral: —, Qwen3.8 27B: 65.9 (#41)
| Benchmark | Codestral | Qwen3.8 27B |
|---|---|---|
| LMArena Text | — | 1441 |
| LMArena Creative Writing | — | 1382 |
| EQ-Bench Creative Writing | — | 1671 |
| LMArena Multi-Turn | — | 1443 |
Frequently asked questions
Is Codestral better than Qwen3.8 27B?
Qwen3.8 27B is the stronger model overall, scoring 45.3 to 30.1 on the Noometry Index. Codestral costs 2.5× less per token, which makes it the better buy when Qwen3.8 27B's lead doesn't matter for your workload.
Which is cheaper, Codestral or Qwen3.8 27B?
Codestral is cheaper. It lists at $0.30 per million input tokens and $0.90 per million output tokens; Qwen3.8 27B lists at $0.99 and $1.49.
Is Codestral or Qwen3.8 27B better for coding?
Qwen3.8 27B scores higher on coding benchmarks: 50.3 versus 26.9 in the Noometry coding category.
Which has the bigger context window?
Qwen3.8 27B does, with 262K tokens against 256K.
How many benchmarks do Codestral and Qwen3.8 27B share?
0 benchmarks have published results for both models. Codestral has 7 scored results on Noometry and Qwen3.8 27B has 28.