Skip to content
English

Switch theme

// Mistral / Magistral / Devstral · Text LLM

Codestral-22B-v0.1: VRAM & local deployment

22.2B parameters. About 17.2 GB at 4-bit and 51.3 GB at BF16 with one 8K-token request.

released 2024-05-29licence: Non-commercialmemory figures are estimates

01Specs

PublisherMistral AI · mistralai
Family / seriesMistral / Magistral / Devstral · Codestral
TypeText LLM
Parameters22.2B Source: counted from official safetensors weights
Context length32K tokens (32,768)max_position_embeddings in config.json
Architecture56 layers · GQA
Released weightsBF16 · 89.0 GBdownload
LicenceMistral Non-Production License · Non-commercialResearch / non-commercial licence: commercial use is not allowed without a separate agreement.
Released2024-05-29 (Hugging Face repo created)

02How much VRAM it needs (estimate)

Total = weights + 10% + 0.5GB overhead + KV cache for one request (FP16 KV). More concurrent users multiply the KV column.

PrecisionWeightsTotal @ 8KTotal @ 32KSmallest GPUMac
BF16 / FP16 (original)44.5 GB 51.3 GB57.0 GB 2 × RTX 5090 (64GB)Mac 128GB
FP8 / INT8 / Q8_023.6 GB 28.4 GB34.0 GB RTX 5090 32GBMac 64GB
4-bit (Q4_K_M / AWQ / GPTQ-Int4)13.5 GB 17.2 GB22.9 GB RTX 4090 24GBMac 36GB

KV cache per request: 8K 1.9 GB · 32K 7.5 GB

Bits per weight: BF16 16, FP8 8.5, 4-bit 4.85. “Fits” = total ≤ 90% of memory; Mac GPU can use ~2/3 of unified memory (≤36GB) or ~3/4 (larger). For MoE, all experts must be in memory, so size by total parameters. Method →

03How to run it

Only runtimes the publisher or the runtime’s official library names for this model.

Which inference engine to choose →

04Official repositories & quantised builds

1 official repo(s) from mistralai.

05Other Mistral / Magistral / Devstral models

All 35 →
ModelParams4-bit totalReleased
LIDstral-Arabic——2026-10-08
Voxtral-Mini-4B-Realtime-Arabic4.43B3.5 GB2026-10-08
Shieldstral-1.0-3B3.85B3.9 GB2026-07-16
Mistral-Medium-3.5-128B127.7B88.6 GB2026-03-31
Mistral-Small-4-119B-2603119B / 6.5B80.0 GB2026-01-23
Voxtral-Mini-4B-Realtime-26024.43B3.5 GB2026-01-21
Devstral-2-123B-Instruct-2512125B86.8 GB2025-11-28
Mistral-Large-3-675B-Instruct-2512675B / 41B451 GB2025-11-28

06Sources

Data checked 2025-07-24. Parameters, context and licence are copied from the official repository; memory figures are our estimates.

Data updated: Self-host open models(UTC+8)