architectures · meituan

LongCat Architecture (Long-Context MoE)

Meituan LongCat sparse MoE architecture specialised for long-context tasks. Employs chunked attention with sliding window and global sink tokens, combined with expert routing tuned for retrieval workloads. Flash, Thinking, and Lite variants share the same core; Flash-Chat is conversation-optimised. Up to 560B total params.

Verify

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	1cba4652b129f30824c08583e25c1bf1cf0483c647e4a779d9d9e835910b3eaa

Re-check this yourself → Source → Full record →

What is recorded

Publisher	meituan
Origin	Recorded as a covered nation under 10 U.S.C. § 4872(f) — the PRC, Russia, Iran or North Korea. The registry did not record which.
Licence	LongCat-Open
Pulls recorded here	156,313

Elsewhere in the registry

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

Registry record as of 2026-06-06. 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.