Model comparison
Codestral vs DeepSeek LLM 67B
Codestral is the stronger model overall, scoring 30.1 to 24.8 on the Noometry Index.
Last verified . 0 shared benchmarks.
Mistral AI
DeepSeekSummary
- The widest gap is in coding, where DeepSeek LLM 67B leads 31.3 to 26.9.
Side by side
| Codestral | DeepSeek LLM 67B | |
|---|---|---|
| Provider | Mistral AI | DeepSeek |
| Noometry Index | 30.1 | 24.8 |
| Released | 2024-05-29 | 2023-11-29 |
| Weights | Open | Open |
| Context window | 256K | — |
| Max output | 4K | — |
| Input $ / M tokens | $0.30 | — |
| Output $ / M tokens | $0.90 | — |
| Results tracked | 7 | 15 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek LLM 67B leads
Codestral: 26.9 (#316), DeepSeek LLM 67B: 31.3 (#273)
| Benchmark | Codestral | DeepSeek LLM 67B |
|---|---|---|
| Aider Polyglot | 11.1% | — |
| BigCodeBench Instruct | 41.8% | — |
| LMArena Coding | — | 1096 |
| BigCodeBench Complete | 52.5% | — |
| ALE-Bench | 137.78 | — |
| HumanEval+ | 73.8% | — |
| MBPP+ | 61.9% | — |
Reasoning Codestral leads
Codestral: 19.1 (#244), DeepSeek LLM 67B: 15.9 (#300)
| Benchmark | Codestral | DeepSeek LLM 67B |
|---|---|---|
| Kagi LLM Benchmark | 32.5% | — |
| Chess Puzzles | — | 0% |
| LMArena Hard Prompts | — | 1070 |
| Epoch Capabilities Index | — | 110.54 |
Math Not comparable
Codestral: —, DeepSeek LLM 67B: 8.1 (#318)
| Benchmark | Codestral | DeepSeek LLM 67B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 0.8% |
| LMArena Math | — | 1108 |
| MATH Level 5 | — | 6.4% |
Knowledge Not comparable
Codestral: —, DeepSeek LLM 67B: 7.1 (#308)
| Benchmark | Codestral | DeepSeek LLM 67B |
|---|---|---|
| GPQA Diamond | — | 24.6% |
Multilingual Not comparable
Codestral: —, DeepSeek LLM 67B: 29.5 (#262)
| Benchmark | Codestral | DeepSeek LLM 67B |
|---|---|---|
| LMArena Non-English | — | 1073 |
| LMArena Chinese | — | 1132 |
Instruction Following Not comparable
Codestral: —, DeepSeek LLM 67B: 55.7 (#272)
| Benchmark | Codestral | DeepSeek LLM 67B |
|---|---|---|
| LMArena Instruction Following | — | 1079 |
Long Context Not comparable
Codestral: —, DeepSeek LLM 67B: 33.1 (#260)
| Benchmark | Codestral | DeepSeek LLM 67B |
|---|---|---|
| LMArena Longer Query | — | 1092 |
Writing & Preference Not comparable
Codestral: —, DeepSeek LLM 67B: 31.8 (#277)
| Benchmark | Codestral | DeepSeek LLM 67B |
|---|---|---|
| LMArena Text | — | 1105 |
| LMArena Creative Writing | — | 1067 |
| LMArena Multi-Turn | — | 1082 |
Frequently asked questions
Is Codestral better than DeepSeek LLM 67B?
Codestral is the stronger model overall, scoring 30.1 to 24.8 on the Noometry Index.
Is Codestral or DeepSeek LLM 67B better for coding?
DeepSeek LLM 67B scores higher on coding benchmarks: 31.3 versus 26.9 in the Noometry coding category.
How many benchmarks do Codestral and DeepSeek LLM 67B share?
0 benchmarks have published results for both models. Codestral has 7 scored results on Noometry and DeepSeek LLM 67B has 15.