models/mistralai-mixtral-8x7b-instruct-v0.1

Mixtral 8x7B v0.1

Sources arena, livebench · Confidence B

params

46.7B

dense

bench sources

mid conf

context

—

max tokens

license

Apache 2.0

permissive

▸Model meta

LiveBench

coding13.3

overall13.3

▸Hardware fit matrix

by gpu tier	Q2_K 17.5 GB	Q4_K_M 23.4 GB	Q5_K_M 29.2 GB	Q6_K 35.0 GB	Q8_0 46.7 GB
8 GB tier rtx 3050 · 4060
12 GB tier rtx 3060 · 4070
16 GB tier rtx 4080 · 4060 ti 16g
24 GB tier rtx 3090 · 4090	GREAT 51 t/s
32 GB tier rtx 5090 · m3 max	GREAT 95 t/s	GREAT 72 t/s
48 GB tier a6000 · m3 max 64	GREAT 31 t/s	GREAT 23 t/s	GREAT 19 t/s	GREAT 16 t/s
80 GB tier h100 · m3 ultra 128	GREAT 104 t/s	GREAT 79 t/s	GREAT 63 t/s	GREAT 53 t/s	GREAT 40 t/s

full fit · production speed tight fit · usabledoesn't fit

Get personalized ranking

Tell us your machine — we'll tell you if this is actually your best pick, or what's better.