llama.cpp
Support for DeepseekV32ForCausalLM with generic DeepSeek Sparse Attention (DSA) implementation
#23346
Merged

Support for DeepseekV32ForCausalLM with generic DeepSeek Sparse Attention (DSA) implementation #23346

fairydreaming
sszymczy llama : support DeepSeek V3.2 model family (with DSA lightning indexer)
efd99b0e
sszymczy model : merge two scale operations into one in DSA lightning indexer …
35b4d811
sszymczy chore : remove unused code
069a7741
fairydreaming fairydreaming requested a review from JohannesGaessler JohannesGaessler 69 days ago
fairydreaming fairydreaming requested a review from CISC CISC 69 days ago
fairydreaming fairydreaming requested a review from ggerganov ggerganov 69 days ago
github-actions github-actions added model
github-actions github-actions added testing
github-actions github-actions added python
github-actions github-actions added ggml
sszymczy Merge remote-tracking branch 'upstream/master' into deepseek-v32-minimal
cb855a34
ggerganov ggerganov assigned ggerganov ggerganov 65 days ago
ggerganov
ggerganov approved these changes on 2026-05-25
CISC
CISC commented on 2026-05-25
am17an
fairydreaming
am17an
fairydreaming model : support NVFP4 in DeepSeek V3.2
4643fdaf
CISC
fairydreaming fairydreaming force pushed from 30fdfe4a to 4643fdaf 60 days ago
sszymczy memory : refactoring TODO
4ce1f300
sszymczy Merge remote-tracking branch 'upstream/master' into deepseek-v32-minimal
101bad43
CISC
CISC approved these changes on 2026-05-28
fairydreaming
fairydreaming
CISC
fairydreaming
CISC
am17an
CISC CISC merged 1f0aa2a6 into master 60 days ago
JohannesGaessler
CISC

Login to write a write a comment.

Login via GitHub

Assignees
Labels
Milestone