Skip to content

Pull requests: antirez/ds4

New pull request New
400 Open 241 Closed
400 Open 241 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

Fix missing Metal object header dependencies
#1037 opened Sep 13, 2026 by IngeniousIdiocy Loading…
ROCm: add DeepSeek V4.1 Flash support for Strix Halo (gfx1151)
#1036 opened Sep 13, 2026 by kyuz0 Contributor Loading…
1
engram: parallelize DeepSeek V4.1 Flash decode reads on macOS
#1035 opened Sep 13, 2026 by Dango233 Loading…
metal: reduce decode synchronization for DeepSeek V4.1 Flash SSD streaming
#1034 opened Sep 13, 2026 by Dango233 Loading…
metal: add opt-in slab residency for DeepSeek V4.1 Flash
#1033 opened Sep 13, 2026 by Dango233 Loading…
CUDA: use selected expert cache for IQ2 SSD prefill
#1031 opened Sep 12, 2026 by Matthley Loading…
Fix CUDA long-context smoke test linking
#1030 opened Sep 12, 2026 by Matthley Loading…
3 tasks done
2
rocm: model weights in host RAM on integrated APUs (Strix Halo)
#1028 opened Sep 12, 2026 by Piega Loading…
Make BPE merging O(n log n) for large CJK pieces
#1027 opened Sep 11, 2026 by wbllcbbg-ai Loading…
2
Enable GLM 5.3 tensor parallelism on the ROCm backend
#1024 opened Sep 10, 2026 by davidcanar Loading…
9
server: opt-in auto-reduction of over-budget image histories (DS4_VISION_KEEP_IMAGES)
#1022 opened Sep 10, 2026 by nazerim Loading…
server: auto-discard reload-loop checkpoints (frontier-contradicted)
#1021 opened Sep 10, 2026 by nazerim Loading…
kvstore: stamp checkpoints with a behavioral tokenizer fingerprint
#1020 opened Sep 10, 2026 by nazerim Loading…
Add OdinLink (odl_tb5) TP transport: --transport odl
#1018 opened Sep 10, 2026 by davidcanar Loading…
server: add --reasoning and --alias CLI options
#1017 opened Sep 9, 2026 by aaa2015 Loading…
Preserve schema-declared JSON types in GLM tool calls
#1016 opened Sep 9, 2026 by babyLegionite Loading…
1
Filter vision images in two separable passes
#1013 opened Sep 9, 2026 by namespaceMarcello Loading…
Server: preserve raw prompts in /v1/completions
#1012 opened Sep 9, 2026 by kyuz0 Contributor Loading…
ROCm: accelerate GLM 5.3 prefill and reduce MTP overhead on gfx1151
#1011 opened Sep 9, 2026 by kyuz0 Contributor Loading…
2
kv-cache: lazy-grow compressed KV caches instead of full-ctx up front
#1010 opened Sep 9, 2026 by evintunador Loading…
Precompute the vision resize tap weights
#1008 opened Sep 9, 2026 by namespaceMarcello Loading…
server: keep SSE streams alive while tool-call output is withheld
#1006 opened Sep 9, 2026 by sscarduzio Loading…
server: support live prefix rewind for GLM-5.3 and preserve common prefix on forks
#1005 opened Sep 9, 2026 by aaa2015 Loading…
Ignore the build outputs tests and benchmarks leave behind
#1004 opened Sep 9, 2026 by daaain Loading…
Reuse DSpark snapshots for immediate session rewinds
#1003 opened Sep 9, 2026 by emilianbold Loading…
2
ProTip! Filter pull requests by the default branch with base:main.