llama.cpp
Support for DeepseekV32ForCausalLM with generic DeepSeek Sparse Attention (DSA) implementation
#23346
Merged
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Overview
Commits
7
Changes
View On
GitHub
Support for DeepseekV32ForCausalLM with generic DeepSeek Sparse Attention (DSA) implementation
#23346
CISC
merged 7 commits into
ggml-org:master
from
fairydreaming:deepseek-v32-minimal
llama : support DeepSeek V3.2 model family (with DSA lightning indexer)
efd99b0e
model : merge two scale operations into one in DSA lightning indexer …
35b4d811
chore : remove unused code
069a7741
fairydreaming
requested a review
from
JohannesGaessler
69 days ago
fairydreaming
requested a review
from
CISC
69 days ago
fairydreaming
requested a review
from
ggerganov
69 days ago
github-actions
added
model
github-actions
added
testing
github-actions
added
python
github-actions
added
ggml
Merge remote-tracking branch 'upstream/master' into deepseek-v32-minimal
cb855a34
ggerganov
assigned
ggerganov
65 days ago
ggerganov
approved these changes on 2026-05-25
CISC
commented on 2026-05-25
model : support NVFP4 in DeepSeek V3.2
4643fdaf
fairydreaming
force pushed
from
30fdfe4a
to
4643fdaf
60 days ago
memory : refactoring TODO
4ce1f300
Merge remote-tracking branch 'upstream/master' into deepseek-v32-minimal
101bad43
CISC
approved these changes on 2026-05-28
CISC
merged
1f0aa2a6
into master
60 days ago
Login to write a write a comment.
Login via GitHub
Reviewers
CISC
ggerganov
JohannesGaessler
Assignees
ggerganov
Labels
model
testing
python
ggml
Milestone
No milestone
Login to write a write a comment.
Login via GitHub