Model comparison
Codestral vs DeepSeek Coder 6.7B
Codestral has enough public results to be ranked (#290); DeepSeek Coder 6.7B does not yet, so treat this comparison as directional.
Last verified . 4 shared benchmarks.
Mistral AI
DeepSeekSummary
- They share 4 benchmarks with published results for both. Codestral scores higher in 0 categories and DeepSeek Coder 6.7B in 1 category; one gap is clear of the uncertainty.
- The widest gap is in coding, where DeepSeek Coder 6.7B leads 35.8 to 27.3.
- The biggest single-benchmark swing is BigCodeBench Complete: 52.5% for Codestral and 43.8% for DeepSeek Coder 6.7B.
- DeepSeek Coder 6.7B has downloadable open weights; the other is API-only.
Side by side
| Codestral | DeepSeek Coder 6.7B | |
|---|---|---|
| Provider | Mistral AI | DeepSeek |
| Noometry Index | 30.6 | 37.6 |
| Released | 2024-05-29 | 2023-11-02 |
| Weights | Proprietary | Open |
| Context window | 256K | — |
| Max output | 8K | — |
| Input $ / M tokens | $0.30 | — |
| Output $ / M tokens | $0.90 | — |
| Results tracked | 7 | 9 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek Coder 6.7B leads
Codestral: 27.3 (#321), DeepSeek Coder 6.7B: 35.8 (#219)
| Benchmark | Codestral | DeepSeek Coder 6.7B |
|---|---|---|
| BigCodeBench Instruct | 41.8% | 35.5% |
| BigCodeBench Complete | 52.5% | 43.8% |
| HumanEval+ | 73.8% | 71.3% |
| MBPP+ | 61.9% | 65.6% |
| Aider Polyglot | 11.1% | — |
| ALE-Bench | 137.78 | — |
Reasoning Not comparable
Codestral: 19.8 (#251), DeepSeek Coder 6.7B: —
| Benchmark | Codestral | DeepSeek Coder 6.7B |
|---|---|---|
| Kagi LLM Benchmark | 32.5% | — |
| Epoch Capabilities Index | — | 89.62 |
| WinoGrande | — | 57.6% |
Math Not comparable
Codestral: —, DeepSeek Coder 6.7B: —
| Benchmark | Codestral | DeepSeek Coder 6.7B |
|---|---|---|
| GSM8K | — | 21.3% |
Knowledge Not comparable
Codestral: —, DeepSeek Coder 6.7B: —
| Benchmark | Codestral | DeepSeek Coder 6.7B |
|---|---|---|
| ARC (AI2) Challenge | — | 36.4% |
| MMLU | — | 36.4% |
Frequently asked questions
Is Codestral better than DeepSeek Coder 6.7B?
Codestral has enough public results to be ranked (#290); DeepSeek Coder 6.7B does not yet, so treat this comparison as directional.
Is Codestral or DeepSeek Coder 6.7B better for coding?
DeepSeek Coder 6.7B scores higher on coding benchmarks: 35.8 versus 27.3 in the Noometry coding category.
How many benchmarks do Codestral and DeepSeek Coder 6.7B share?
4 benchmarks have published results for both models. Codestral has 7 scored results on Noometry and DeepSeek Coder 6.7B has 9.