PR #16477 Add AfmoeForCausalLM support

Add AfmoeForCausalLM support #16477

CISC merged 10 commits into ggml-org:master from bartowski1182:master

github-actions added python

bartowski1182 force pushed 38 days ago

github-actions added model

Add AFMOE model support

3fd69c59

bartowski1182 force pushed to 3fd69c59 37 days ago

Update to vocab

93a2fb46

Add model sizing

7c5d7184

bartowski1182 marked this pull request as ready for review 30 days ago

bartowski1182 requested a review from

CISC 30 days ago

bartowski1182 requested a review from

ggerganov 30 days ago

Undo Rope change for ARCEE model

13aaafe6

CISC commented on 2025-11-13

Address review comments

34dc2a37

bartowski1182 commented on 2025-11-13

ngxson commented on 2025-11-13

Update modeling code is_sliding -> use_rope, replace hard-coded logic

1b9558f8

Fix AFMOE tokenizer

e41a5bd4

CISC commented on 2025-11-13

bartowski1182 commented on 2025-11-13

CISC commented on 2025-11-13

Update convert_hf_to_gguf.py

e27c3f1d

Update convert_hf_to_gguf.py

684aeada

Update AFMoE tokenizer class identification to be more unique

ddddf8d7

CISC approved these changes on 2025-11-14

ngxson approved these changes on 2025-11-14

CISC merged e1fcf8b0 into master 29 days ago

Reviewers

ngxson

CISC

ggerganov

Assignees

No one assigned

Labels

model python

Milestone

No milestone

llama.cpp Add AfmoeForCausalLM support #16477 Merged

Add AfmoeForCausalLM support #16477

llama.cpp
Add AfmoeForCausalLM support
#16477

Merged