← Back to dashboard
Mistral·Code — weighted competency·Tue 2 Jun 2026

Codestral 3

Codestral 3 scores 72/100 competency for Code at $1.80.

Low confidenceOpen weights
Competency
72/100
category rank #6
Cost
$1.80/ 1M tokens
blended
Speed
82/100
Fast
Accuracy
3.5% halluc.
High confidence

Score history.

Weighted Code competency · last 12 weeks · weekly snapshot.

72/100
— flat
72

What it's actually good at.

Score on each business-task subset that goes into the weighted competency.

Test Generation10% of category weight
80
Code Generation40% of category weight
68
Review & Bug Fixing30% of category weight
72
Full-Repo Understanding20% of category weight
76

Best for. Watch out for.

✓ Best for
  • Strong at Code Generation for Mistral.
  • Strong at Review & Bug Fixing for Mistral.
⚠ Watch out for
  • Validate fit for your specific workload before committing.

Pricing.

Published usage pricing. Figures refresh daily from our pricing sources.

$1.80

Usage pricing

$1.80/1M tok
Input
$1.80 / 1M tokens
Output
$5.40 / 1M tokens

Where it sits.

Position against the rest of the category. Ranked by weighted competency.

RankModelScoreCostSpeedΔ vs Codestral
01GPT-5 Code79$14.0074 / 100+7
02DeepSeek V3.1 Coder64$0.676 / 100-8
03Qwen 3 Coder48$1.4078 / 100-24