Mistral Codestral — Benchmark Results
Mistral Overall Score: 60/100
Performance Scores
Overall
60
Rank #28 of 45 — Top 38%
SWE-bench
54
Rank #28 of 45 — Top 38%
LiveCodeBench
64
Rank #26 of 45 — Top 42%
HumanEval
82
Rank #27 of 45 — Top 40%
BigCodeBench
44
Rank #29 of 45 — Top 36%
Strengths & Weaknesses
Strengths
- Code-specialized
- Very cheap
Weaknesses
- Narrow focus
Compare with Similar-Priced Models
| Model | Overall Score | Input $/M |
|---|---|---|
| Mistral Codestral | 60 | $0.300 |
| Claude 3.5 Haiku | 52 | $0.800 |
| Claude 3 Haiku | 45 | $0.250 |
| GPT-4o mini | 58 | $0.150 |
| GPT-3.5 Turbo | 40 | $0.500 |
| OpenAI o1-mini | 70 | $1.10 |