Codestral — Local AI Model by Mistral AI

作者： Jakub Rusinowski · 最后更新： 2024年5月29日

Superseded model. Codestral has been superseded by Devstral. This page is kept for reference; the newer family is a better starting point. View Devstral →

Mistral专用的代码生成模型，在覆盖80+编程语言的庞大编程语料库上训练。擅长填充中间（FIM）补全——这是IDE自动补全的核心技术。原生集成VS Code的continue.dev和Cursor。

Licence

Licence	What it permits	Applies to
`MNPL-0.1`	Research / non-commercial only Research / non-commercial only — this licence does NOT permit shipping a commercial product.	Codestral 22B

Hardware Requirements

Codestral 22B Min 14 GB VRAM · Q4_K_M · 32,768 ctx · ollama run codestral:22b

Recommended GPU

The cheapest GPU that runs Codestral locally (min 14 GB VRAM) is the AMD Radeon RX 9060 XT 16GB (16 GB).

联盟营销声明: 本页部分链接为联盟推广链接——如果你通过它们购买，LLM Configurator 可能会获得佣金，而你无需支付任何额外费用。作为亚马逊联盟成员（Amazon Associate），LLM Configurator 会从符合条件的购买中获得收益。

AMD Radeon RX 9060 XT 16GB

16 GB VRAM · 160 W board power

2026年价格波动较大——请以当前商品页价格为准。

在亚马逊查看价格

How to Run Locally

Install Ollama then run: ollama run codestral:22b

Minimum VRAM: 14 GB. For best results use Q4_K_M quantization.

Codestral — Frequently Asked Questions

How much VRAM does Codestral need?

Codestral needs about 14 GB VRAM at Q4_K_M quantization for its smallest variant. Variants: Codestral 22B (14 GB, Q4_K_M). On Apple Silicon, unified memory counts toward this requirement.

Can I run Codestral on an RTX 4090 (24 GB)?

Yes — Codestral runs on an RTX 4090 (24 GB) and other 24 GB cards such as the RTX 3090. Smaller variants also fit comfortably on 8–16 GB GPUs at Q4_K_M.

What quantization should I use for Codestral?

Q4_K_M is the best balance of quality and VRAM for Codestral in most cases. Choose Q8_0 for near-lossless quality if you have spare VRAM, or smaller quants (Q3/Q2) only when memory is tight.

How do I run Codestral with Ollama?

Install Ollama, then run: ollama run codestral:22b. This downloads Codestral and starts a local, OpenAI-compatible endpoint — no internet connection is needed after the initial download.