Mistral AI

Mixtral 8x7B

46,7B

Parameters

24,6 Go

Q4_K_M file

≥ 26,1 Go

Recommended VRAM

32 768

Max context

Not measured yet

No card has run the TokenBench protocol on this model yet. The current protocol covers a selection of models; benchmarking any catalog model from the app is coming soon.

Measure my card →

Which cards should it fit on?

Cards on the site whose video memory covers the recommended 26,1 Go (file + 4,000-token context). An estimate, not a measurement. Click a card for the detailed estimate.

Technical sheet

Published: 2023-12-10
Architecture: 32 layers · 8 KV heads · dimension 128
Attention: Standard: every layer keeps the whole context in memory.
Context cache (f16): ≈ 125 MB per 1,000 tokens · 4,0 Go for 32,768 tokens
GGUF file: TheBloke/Mixtral-8x7B-Instruct-v0.1-GGUF · mixtral-8x7b-instruct-v0.1.Q4_K_M.gguf Download (24,6 Go)
Official repo: mistralai/Mixtral-8x7B-Instruct-v0.1

Public Hugging Face data (official config.json, GGUF repo). The context cache only counts full-attention layers.