-
-
Notifications
You must be signed in to change notification settings - Fork 44
Pull requests: charlie12345/ROCmFPX
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
[Draft/RFC] Port DFlash2 support from llama.cpp PR #27342
conversion
model
server
#90
opened Aug 20, 2026 by
serrynaimo
•
Draft
common: backport upstream #26252 qwen3 tool-call parser + peg-builder API
documentation
Improvements or additions to documentation
jinja parser
server
#88
opened Aug 19, 2026 by
jtstothard
Loading…
Fix Nemotron-H MoE: load NVFP4 expert scale2 tensors + remove double-scale
model
#87
opened Aug 18, 2026 by
poad42
Loading…
server: per-request reasoning_budget_message and exact cache round-trip with a wrap-up message
server
#80
opened Aug 17, 2026 by
pugant
Loading…
model: add cohere2moe architecture (Cohere North-Mini-Code)
conversion
model
#53
opened Aug 5, 2026 by
kingjones30
Contributor
Loading…
Add Hy3 IFP2 runtime, adaptive MoE, and SSD MTP cache
conversion
CUDA
documentation
Improvements or additions to documentation
ggml
server
testing
Document released model recipe contracts
devops
documentation
Improvements or additions to documentation
ggml
#21
opened Jul 9, 2026 by
ciru-ai
Contributor
Loading…
Incorporate SSD Caching from CachyLLama (Experimental)
devops
documentation
Improvements or additions to documentation
server
#20
opened Jul 6, 2026 by
3rdIteration
•
Draft
ProTip!
Updated in the last three days: updated:>2026-08-17.