models · mistralai

Mixtral-8x22B-Instruct

Sparse Mixture-of-Experts (8x22B, 39B active). Strong throughput-to-quality ratio for self-hosted inference.

What this registry has checked

Quarantine. We hold the publisher’s own published hash for these weights and have recorded it against this entry. We have not downloaded and hashed the weights ourselves, so the hash is the publisher’s claim, not our measurement.

Weight file SHA-256	94863d0958fe623f211594f0812e59faad5a40fd85bfb4b975e9b63f644ee0e9
Method	hf-published-sha256

Re-check this yourself → Source → Full record →

What is recorded

Publisher	mistralai
Origin	Recorded as allied origin.
Licence	Apache-2.0
Parameters	140,630,071,296 (140.6B)
Context window	64K
Modality	text+file->text
Architecture	mixtral
Layers	56
Hidden size	6144
Attention	GQA (48 heads, 8 KV)
Native precision	BF16
Pulls recorded here	410,016
Downloads reported upstream	24,157

Reported scores

MMLU-Pro	53.7
GPQA Diamond	33.2
MATH-500	54.5
LiveCodeBench	14.8
Humanity's Last Exam	4

Reported by a third party and recorded here, not re-run by us. Source: artificialanalysis.ai/api/v2.

Elsewhere in the registry

All models → Browse with filters → What will run on my hardware? →

Registry record as of 2026-06-11. Identity, licence and parameter count are properties of a release and do not move; tier can, and the verify link above re-checks it against the live registry rather than this page.