-
Notifications
You must be signed in to change notification settings - Fork 1.5k
Pull requests: antirez/ds4
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
server: surface unclosed thinking as reasoning_content, not content
#524
opened Jul 9, 2026 by
stefanwichmann
Loading…
hy3: Metal decode path + shared gate/up fusion
#523
opened Jul 9, 2026 by
audreyt
Contributor
Loading…
3 of 5 tasks
glm: fix streaming prefill failures for real-size prompts
#520
opened Jul 8, 2026 by
andreaborio
Loading…
Server: add --parallel N interleaved sessions (default 1)
#516
opened Jul 7, 2026 by
iCreil
Loading…
Engine: share prefill/decode work buffers across sessions
#515
opened Jul 7, 2026 by
iCreil
Loading…
cuda: fix uninitialized IQ2 dequant LUTs in batch MoE expert-tile kernels (n_embd > 4096)
#513
opened Jul 7, 2026 by
giannisanni
Loading…
port: Windows native ROCm build for AMD Strix Halo (gfx1151)
#512
opened Jul 7, 2026 by
paperlinguist
•
Draft
Fix CPU backend abort on DeepSeek V4 PRO: use DS4_N_OUT_GROUP for attention-output grouping
#511
opened Jul 6, 2026 by
mrighini
Loading…
CPU backend: SSD streaming for routed experts (safe CPU inference on macOS)
#508
opened Jul 6, 2026 by
Tatlatat
Loading…
CUDA streaming: per-size-class expert caches, keep cache warm across prefills
#504
opened Jul 5, 2026 by
iCreil
Loading…
Reconcile DSpark runtime (#480) + B2 rejection sampling/adaptive block sizing (#482)
#502
opened Jul 5, 2026 by
stephenlthorn
Loading…
Opt-in mlock of non-routed weights for SSD streaming (DS4_MLOCK_NONROUTED)
#499
opened Jul 4, 2026 by
hi-pauls
Loading…
6 tasks done
rocm: use numerically stable SiLU to avoid NaN on large-magnitude gate
#498
opened Jul 4, 2026 by
Flowers-of-Romance
Loading…
cuda: run streaming selected load for single-token ffn batch encodes
#497
opened Jul 4, 2026 by
giannisanni
Loading…
Server: fix agent-loop cache misses, add cancellation, observability, and robustness fixes
#489
opened Jul 2, 2026 by
elkaix
Loading…
cuda: enable streaming auto cache (implement recommended_working_set_size)
#488
opened Jul 2, 2026 by
riccardo-galbani
Loading…
cuda: fall back to pinned host memory when the model arena runs out of VRAM
#487
opened Jul 2, 2026 by
riccardo-galbani
Loading…
DSpark B2 rejection sampling + adaptive block sizing
#482
opened Jun 30, 2026 by
machiabeli
Loading…
feat: add headless browser support with curl fallback for web tools
#479
opened Jun 29, 2026 by
J3rr1ck
Loading…
Previous Next
ProTip!
Filter pull requests by the default branch with base:main.