Model comparison
Codestral vs DeepSeek-V3.2-Speciale
DeepSeek-V3.2-Speciale is the stronger model overall, scoring 39.1 to 30.1 on the Noometry Index. Codestral costs 1.9× less per token, which makes it the better buy when DeepSeek-V3.2-Speciale's lead doesn't matter for your workload.
Last verified . 0 shared benchmarks.
Mistral AI
DeepSeekSummary
- The widest gap is in reasoning, where DeepSeek-V3.2-Speciale leads 32.1 to 19.1.
- Codestral is cheaper at $0.30 / $0.90 per million input/output tokens, against $0.58 / $1.68 for DeepSeek-V3.2-Speciale.
- Codestral accepts more context: 256K tokens versus 128K.
Side by side
| Codestral | DeepSeek-V3.2-Speciale | |
|---|---|---|
| Provider | Mistral AI | DeepSeek |
| Noometry Index | 30.1 | 39.1 |
| Released | 2024-05-29 | 2025-12-01 |
| Weights | Open | Open |
| Context window | 256K | 128K |
| Max output | 4K | 128K |
| Input $ / M tokens | $0.30 | $0.58 |
| Output $ / M tokens | $0.90 | $1.68 |
| Results tracked | 7 | 3 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.2-Speciale leads
Codestral: 26.9 (#316), DeepSeek-V3.2-Speciale: 39.9 (#136)
| Benchmark | Codestral | DeepSeek-V3.2-Speciale |
|---|---|---|
| Aider Polyglot | 11.1% | — |
| WeirdML | — | 46.7% |
| BigCodeBench Instruct | 41.8% | — |
| BigCodeBench Complete | 52.5% | — |
| ALE-Bench | 137.78 | — |
| HumanEval+ | 73.8% | — |
| MBPP+ | 61.9% | — |
Reasoning DeepSeek-V3.2-Speciale leads
Codestral: 19.1 (#244), DeepSeek-V3.2-Speciale: 32.1 (#65)
| Benchmark | Codestral | DeepSeek-V3.2-Speciale |
|---|---|---|
| SimpleBench | — | 52.6% |
| Kagi LLM Benchmark | 32.5% | — |
Writing & Preference Not comparable
Codestral: —, DeepSeek-V3.2-Speciale: 46.1 (#217)
| Benchmark | Codestral | DeepSeek-V3.2-Speciale |
|---|---|---|
| EQ-Bench Creative Writing | — | 1276 |
Frequently asked questions
Is Codestral better than DeepSeek-V3.2-Speciale?
DeepSeek-V3.2-Speciale is the stronger model overall, scoring 39.1 to 30.1 on the Noometry Index. Codestral costs 1.9× less per token, which makes it the better buy when DeepSeek-V3.2-Speciale's lead doesn't matter for your workload.
Which is cheaper, Codestral or DeepSeek-V3.2-Speciale?
Codestral is cheaper. It lists at $0.30 per million input tokens and $0.90 per million output tokens; DeepSeek-V3.2-Speciale lists at $0.58 and $1.68.
Is Codestral or DeepSeek-V3.2-Speciale better for coding?
DeepSeek-V3.2-Speciale scores higher on coding benchmarks: 39.9 versus 26.9 in the Noometry coding category.
Which has the bigger context window?
Codestral does, with 256K tokens against 128K.
How many benchmarks do Codestral and DeepSeek-V3.2-Speciale share?
0 benchmarks have published results for both models. Codestral has 7 scored results on Noometry and DeepSeek-V3.2-Speciale has 3.