architectures · zhipu-ai

GLM Architecture (General Language Model)

Zhipu AI General Language Model architecture. Autoregressive blank infilling pre-training with bidirectional context via 2D positional encodings. GLM-4.x through GLM-5.x series use a scaled MoE variant with Flash Attention and a 200K+ token context window via ROPE. Achieves SOTA on Chinese+English benchmarks.

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	51fbcf77f176031372f41e4e39068b3e0ec81286a94a02547f17e1b5edff2c6c

Re-check this yourself → Source → Full record →

What is recorded

Publisher	zhipu-ai
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	MIT
Pulls recorded here	148,831

Elsewhere in the registry

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

Registry record as of 2026-08-14. 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.