Skip to content

Pull requests: antirez/ds4

New pull request New
66 Open 85 Closed
66 Open 85 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Add --kv-cache turbo3 (TurboQuant+ 3-bit KV cache, CUDA + Metal)
#243 opened May 24, 2026 by TheTom Loading…
5
ds4_server: check if client disconnected to abort inference early
#242 opened May 24, 2026 by confiden Loading…
agent: resolve runtime assets from binary directory
#241 opened May 24, 2026 by dbader Loading…
ds4-server: watchdog thread + decode-loop SSE keepalive
#238 opened May 24, 2026 by Allen091080 Loading…
3 tasks
1
Validate CLI seed and server port arguments
#234 opened May 23, 2026 by FabioMalpezzi Loading…
Continuous batching
#230 opened May 23, 2026 by Chida82 Loading…
Fix SSE keepalive to use wall-clock timer, not progress callback
#227 opened May 22, 2026 by ridhinva Loading…
1
download_model.sh: auto-resume interrupted downloads
#225 opened May 22, 2026 by datag00n Loading…
Decouple SSE prefill keepalive from progress callback
#224 opened May 22, 2026 by njbrake Loading…
1
Rename AGENT.md to AGENTS.md, add CLAUDE.md symlink
#223 opened May 22, 2026 by njbrake Loading…
rocm: sync branch with main
#219 opened May 21, 2026 by 24601 • Draft
6
fix: handle CR in agent yes/no prompts
#215 opened May 21, 2026 by arsis-dev Loading…
[codex] Add bitlift sidecar runtime and eval tools
#212 opened May 21, 2026 by Chedrian07 • Draft
Fix wording in agent feature description
#207 opened May 19, 2026 by Chida82 Loading…
mtp: make speculation disable skip draft work
#206 opened May 19, 2026 by abhay Loading…
bench: add MTP options to ds4-bench
#205 opened May 19, 2026 by tnhnyc Loading…
test: run logprob vectors through quality path
#204 opened May 19, 2026 by abhay Loading…
fix: sanitize user content against special-token injection in chat prompt rendering
#197 opened May 19, 2026 by mvanhorn Loading…
1
docs: explain why README recommends openai-completions for Pi agent client
#196 opened May 19, 2026 by mvanhorn Loading…
1
opt-in sampler-level repetition guard
#195 opened May 19, 2026 by ayourtch Contributor Loading…
3
Fast CUDA backend: vendored mmq matmul, VMM weight layout, speculative-decoding plumbing
#187 opened May 18, 2026 by Entrpi Loading…
Compressed KV disk cache via streaming LZ4
#186 opened May 18, 2026 by jasontitus Loading…
1
rocm: enable wmma indexer support
#180 opened May 17, 2026 by alantsev Contributor Loading…
4
[codex] Fix CUDA object rebuild on config changes
#178 opened May 17, 2026 by audreyt Contributor • Draft
fix(kv-cache): boost eviction score for cold/evict/shutdown checkpoints
#177 opened May 17, 2026 by unsaltedbutter-ai Contributor Loading…
4 tasks
1
ProTip! Add no:assignee to see everything that’s not assigned.