Wednesday, October 28, 2026 🏢 AI Companies Hub RSS About Contact Admin

Codestral 25B

Specialized Code Transformer

Mistral AI 📄 Mistral Non-Production / Commercial API 🌍 France (Europe) 🗓️ Effective Date: 29.05.2024

Dedicated programming and code synthesis model fluent across 80+ programming languages with fill-in-the-middle support.

Model Size / Parameters 25B Dense Architecture: Specialized Code Transformer

Context Window (Input) 256k tokens Token Capacity

Max Output Tokens 32k tokens Single Generation Limit

Supported Modalities Text, Code Native Multimodal Ingestion

License & Access Mistral Non-Production / Commercial API Open Weights vs. API

Release Date May 2024 Effective Date: 29.05.2024

Verified Benchmark Results

Standardized academic and industry evaluation benchmarks (Effective Date: 29.05.2024).

MMLU-Pro / 5-Shot General Intellect

78.5%

SWE-bench Verified Software Engineering

41.2%

MATH 500 / AIME Mathematical Reasoning

79.0%

GPQA Diamond Doctoral Science

42.0%

HumanEval / Python Algorithm Synthesis

81.1%

Official API token pricing and self-hosted licensing terms.

Input Pricing (Prompt)

$0.20 / 1M tokens

Per 1 million processed tokens

Output Pricing (Generation)

$0.60 / 1M tokens

Per 1 million generated tokens

Verified Datasheet: Source: Mistral AI Research Documentation • Effective Date: 29.05.2024

3 weeks ago