Skip to content

Pull requests: antirez/ds4

New pull request New
198 Open 148 Closed
198 Open 148 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

Fixes thinking levels for pi.dev setup in README.md
#561 opened Jul 13, 2026 by daniel-j-h Loading…
Update AGENT.md to match the current engine state
#560 opened Jul 13, 2026 by Cioscos Loading…
ROCm: run on discrete RDNA4 (gfx1201) — opt-in WMMA fallback + configurable VRAM reserve
#558 opened Jul 13, 2026 by cm999club Loading…
1
feat: implementing vulkan backend
#557 opened Jul 13, 2026 by Farenheith • Draft
2
Add ds4-doctor self-diagnostic binary
#556 opened Jul 12, 2026 by AnandSundar Loading…
Optimize Metal prefill and decode on Apple Silicon Pre-M5
#555 opened Jul 12, 2026 by ivanfioravanti Contributor Loading…
2
glm5.2: streamed prefill fails when the boot model map does not cover routed experts
#553 opened Jul 11, 2026 by sk8erboi17 Loading…
Fixes compilation and linking for strix halo rocm target
#552 opened Jul 11, 2026 by daniel-j-h Loading…
ds4-agent: fixes missing --rocm backend argument parsing
#551 opened Jul 11, 2026 by daniel-j-h Loading…
feat(kvstore): decay eviction freshness and split hit half-life by role
#550 opened Jul 11, 2026 by unsaltedbutter-ai Contributor Loading…
Bound tokenizer merges count to INT32_MAX like the tokens table
#548 opened Jul 11, 2026 by gigioneggiando Loading…
Make tool-output validation O(n) instead of O(n^2) (pre-auth CPU DoS)
#547 opened Jul 11, 2026 by gigioneggiando Loading…
Reject oversized snapshot token_count before allocating (distributed DoS)
#546 opened Jul 11, 2026 by gigioneggiando Loading…
Bound GGUF metadata/tensor counts by file size before allocating (DoS)
#545 opened Jul 11, 2026 by gigioneggiando Loading…
Cap file-declared lengths in agent KV cache reads (memory DoS)
#544 opened Jul 11, 2026 by gigioneggiando Loading…
Cross-check tensor data size against shape in FP8/FP4 dequant (OOB read)
#543 opened Jul 11, 2026 by gigioneggiando Loading…
Bound file-declared length prefixes in the model loader (heap overflow)
#542 opened Jul 11, 2026 by gigioneggiando Loading…
Handle NaN in json_int() (undefined double->int cast)
#541 opened Jul 11, 2026 by gigioneggiando Loading…
Unlink bash tool output temp file on job cleanup
#540 opened Jul 11, 2026 by gigioneggiando Loading…
Fix pre-auth double-free in JSON request parser
#539 opened Jul 11, 2026 by gigioneggiando Loading…
Fix gguf-tools quality-score linking
#537 opened Jul 11, 2026 by ivanfioravanti Contributor Loading…
1
Hy3 (295B): CUDA backend support, GQA attention path
#535 opened Jul 11, 2026 by giannisanni Loading…
metal: stop expert-miss readahead racing the pread pool (+13% GLM streaming decode)
#533 opened Jul 10, 2026 by jasontitus Loading…
1
Batched decode: co-decode N live sessions in one GPU pass (opt-in)
#530 opened Jul 10, 2026 by iCreil • Draft
1
Enhance CPU performance with AVX512/AMX SIMD, TBB integration, and NUMA support
#529 opened Jul 10, 2026 by skwerner • Draft
ProTip! no:milestone will show everything without a milestone.