Mistral Models — Hardware Requirements

15 Mistral models from Mistral AI and the community, from the smallest that runs in 0.3 GB of VRAM up to 24.0B parameters. Every row links to full quantization tables and GPU compatibility.

All Mistral Models by Size

ModelParamsContext
PicoMistral 23M24M1K
Mistral 7B Instruct v0.17B33K
OpenHermes 2.5 Mistral 7B7B33K
Mistral 7B v0.17B33K
BioMistral 7B7B33K
Mistral 7B Instruct v0.27.2B33K
Mistral 7B v0.27.2B33K
Mistral 7B Instruct v0.37.2B33K
Mistral Nemo Instruct 240712.2B131K
Mistral Small Instruct 240922.2B33K
Mistral Small 24B Instruct 250123.6B33K
Magistral Small 250623.6B41K
Mistral Small 24B Instruct 2501 Quantized.w8a823.6B33K
Mistral Small 3.2 24B Instruct 250624.0B131K
Mistral Small 3.2 24B Qiskit24.0B131K

How Mistral Compares — Benchmark Rating

Mistral 7B v0.1 is the highest-rated Mistral model with an overall benchmark rating of 57.6/100 — #19 among 75 open models. The top proprietary model, GPT 5.5, scores 88.8. Click a model to see its full benchmark breakdown.

GPT 5.5 · proprietary88.8
Claude Opus 4.7 · proprietary87.6
Claude Fable 5 · proprietary86.6
GPT 5.4 · proprietary86.6
Claude Opus 4.8 · proprietary84.4
Composite of normalized public benchmark scores (methodology) · Mistral · other models

Frequently Asked Questions

How much VRAM do I need to run a Mistral model?
The smallest Mistral model, PicoMistral 23M, runs from 0.3 GB of VRAM at an aggressive quantization. Larger family members need proportionally more — see the table above for every model.
Which Mistral models can I run on a 16 GB GPU?
15 of 15 Mistral models fit in 16 GB of VRAM at some quantization, including Mistral 7B Instruct v0.2, Mistral 7B Instruct v0.3, Mistral Nemo Instruct 2407.
What is the most popular Mistral model to run locally?
Mistral 7B Instruct v0.2 is the most downloaded Mistral model in local-friendly quantized formats. It runs from 3.6 GB of VRAM.
How do Mistral models score on benchmarks?
Mistral 7B v0.1 leads the family with an overall benchmark rating of 57.6/100, ranking #19 among 75 open models, while the top proprietary model, GPT 5.5, scores 88.8. See the comparison chart above for the full standings.