Commit History

Jul 21

06:27
Calibrate MTP layer with runtime-aware AWQ⁠

h4x3rotab

Verified signature
e42e1ae

Jul 8

05:13
Model card: link Chinese documentation on ModelScope

h4x3rotab

Verified signature
438e39c

Jul 7

09:21
Model card: document sampling defaults fix

h4x3rotab

Verified signature
7f42bd8
09:20
Add generation_config.json (temperature=1.0, top_p=0.95, ZAI defaults)⁠

h4x3rotab

Verified signature
427b2cb

Jun 22

07:44
Drop internal index.json fix note from model card

h4x3rotab

Verified signature
8dc3a42

Jun 19

16:46
Model card: trim the 2026-06-19 fix note to the essentials

h4x3rotab

Verified signature
3cd8b80
10:37
Model card: note 2026-06-19 MTP/index fix (re-pull index.json for working speculative decoding)

h4x3rotab

Verified signature
d975c19
10:33
Fix index.json: map layer-78 (MTP/nextn) to shard model-00041 — restores EAGLE speculative decoding (was orphaned/unindexed -> 0.1% accept)

h4x3rotab

Verified signature
d67b0a0

Jun 18

22:02
Update model card: friendly factual lead, H100 KV note, tightened wording

h4x3rotab

Verified signature
1fc4eba
21:36
Card: replace em-dash table placeholders with n/a

h4x3rotab

Verified signature
4d3e051
21:35
Card: plain professional prose (drop pull-quote banners), restore published reference benchmarks

h4x3rotab

Verified signature
6ec26bf
21:30
Card: prominent 1M-context + no-loss claims; clarify param count (4-bit packing, same ~750B); align with official GLM-5.2 without internal refs

h4x3rotab

Verified signature
be158a3
21:20
Restructure model card: quantization summary on top, original GLM-5.2 card appended

h4x3rotab

Verified signature
3ab9e30
21:11
Add model card

h4x3rotab

Verified signature
7fbf93e
20:35
Upload model-00041-of-00041.safetensors with huggingface_hub

h4x3rotab

Verified signature
cc026cc
09:28
Add files using upload-large-folder tool

h4x3rotab

Verified signature
f4af1c5
09:22
initial commit

h4x3rotab

Verified signature
0890d4b