-
Notifications
You must be signed in to change notification settings - Fork 24
Pull requests: hw-native-sys/pypto-serving
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
perf(serving): batch DeepSeek V4 cache metadata preparation
#216
opened Sep 3, 2026 by
high-cloud
Contributor
Loading…
Add DSpark target-model serving (deepseek_v4_flash_dspark)
#213
opened Aug 30, 2026 by
ndleslx
Collaborator
Loading…
fix(deepseek): reserve rings before KV sizing
#209
opened Aug 29, 2026 by
vegetabledoww
Contributor
Loading…
perf(deepseek): build decode block tables without a temporary per row
#206
opened Aug 27, 2026 by
lterrac
Collaborator
Loading…
Revert "perf(deepseek): reduce fused MTP decode host transfers"
#186
opened Aug 24, 2026 by
ndleslx
Collaborator
Loading…
feat(deepseek): add optional MoE expert statistics dump
#185
opened Aug 23, 2026 by
high-cloud
Contributor
•
Draft
feat(deepseek): expand continuous batching across fixed kernels
#145
opened Aug 7, 2026 by
hashiqiqixian
Loading…
feat(skills): add DeepSeek V4 online SA_PROFILE perf skill
#121
opened Jul 28, 2026 by
sunghajung6688
Contributor
Loading…
[feature] TQ serving: add TurboQuant prefill/decode branch alongside …
#114
opened Jul 24, 2026 by
sunghajung6688
Contributor
Loading…
ci: prevent orphaned DeepSeek V4 engines on test interruption
#86
opened Jul 14, 2026 by
luohuan19
Contributor
Loading…
decouple qwen3 serving from lib contract
#85
opened Jul 14, 2026 by
bumble0918
Collaborator
Loading…
perf(qwen3): compile prefill/decode/greedy straight from the signature
#73
opened Jul 13, 2026 by
lyfne123
Contributor
Loading…
perf(qwen3): skip redundant decode d2h/host copies in greedy path
#71
opened Jul 13, 2026 by
lyfne123
Contributor
Loading…
Add serving progress and adjust deepseek weight load
#63
opened Jul 9, 2026 by
superxf
Collaborator
Loading…
Previous Next
ProTip!
Adding no:label will show everything without a label.