model : add hunyuan moe #14425
model : add hunyuan moe
f5d8a227
CISC
commented
on 2025-06-27
tokenizer ok
38acf7fe
fix tensor name
35591a9a
ngxson
commented
on 2025-06-27
cgraph init
cb1f9f2d
chat template
51886a47
wip
cff16cc6
almost working
5e78e887
skip embed, fix bos
d2195807
Merge remote-tracking branch 'other/xsn/hunyuan-moe' into hunyuan
616f4c7a
cleanup
0fd39308
yarn scaling
b19ecae2
cleanup
245db159
correct rope type
3920faa2
failed token fix
8fd547bd
Merge remote-tracking branch 'other/xsn/hunyuan-moe' into hunyuan
34cc679a
ntk alpha freq_base
4d66bdc0
tokenization working
b20bd263
Merge remote-tracking branch 'other/xsn/hunyuan-moe' into hunyuan
99d9e946
cleanup and pr changes
1221d944
vocab_size sanity check
5471f5ac
ntk alpha generic
46c8b70c
Merge pull request #26 from kooshi/hunyuan
443ec9be
ngxson
commented
on 2025-06-30
Update convert_hf_to_gguf.py
251e78a4
ngxson
commented
on 2025-06-30
ngxson
commented
on 2025-06-30
Apply suggestions from code review
06cab8f9
ngxson
marked this pull request as ready for review 165 days ago
Merge branch 'master' into xsn/hunyuan-moe
5cfc73b8
fix regression
2d56a296
CISC
approved these changes
on 2025-06-30
fix style
e5fe0892
kzjeef
commented
on 2025-07-04
ggerganov
merged
8f22dc0a
into master 158 days ago
Assignees
No one assigned
Login to write a write a comment.
Login via GitHub