-
Notifications
You must be signed in to change notification settings - Fork 355
Pull requests: ModelTC/LightLLM
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
feat: prioritize short chunked prefill requests
#1515
opened Aug 26, 2026 by
shihaobai
Collaborator
Loading…
fix(pd): reject requests when master is at capacity
#1514
opened Aug 26, 2026 by
sufubao
Collaborator
Loading…
perf(mtp): optimize Qwen3.5 GDN verification kernel
#1513
opened Aug 26, 2026 by
sufubao
Collaborator
Loading…
Optimize draft logits communication with sparse candidates
#1512
opened Aug 26, 2026 by
hiworldwzj
Collaborator
Loading…
feat(pd): expose stage and node load metrics
#1509
opened Aug 26, 2026 by
sufubao
Collaborator
Loading…
fix(pd): guard remote_agents dict with a lock to fix NIXL concurrent-add race (GH-1470)
#1505
opened Aug 26, 2026 by
Ashfaqbs
Loading…
docs: CONTRIBUTING.md references yapf/Google style guide, but pre-commit actually runs black + flake8
#1504
opened Aug 25, 2026 by
PARINAYKAPOOR
Loading…
fix: setup.py python_requires stale at >=3.9.16, docs require 3.10
#1503
opened Aug 25, 2026 by
PARINAYKAPOOR
Loading…
fix(pd): account for KV transfer pages in mem_fraction
#1495
opened Aug 25, 2026 by
sufubao
Collaborator
Loading…
perf(pd): compact token forwarding packets
#1487
opened Aug 24, 2026 by
WANDY666
Contributor
Loading…
perf(pd): omit unused output logprob metadata
#1486
opened Aug 24, 2026 by
WANDY666
Contributor
Loading…
perf(pd): reduce disaggregated serving latency
#1481
opened Aug 24, 2026 by
sufubao
Collaborator
Loading…
fix(pd): fail requests when assigned nodes disconnect
#1473
opened Aug 20, 2026 by
sufubao
Collaborator
Loading…
fix: apply x_stride1 in the rmsnorm normalize pass
#1468
opened Aug 17, 2026 by
truong-v
Contributor
Loading…
fix(security): stop pickling the multinode ip handshake (GH-1413)
#1458
opened Aug 11, 2026 by
DivyaNarahari97
Loading…
fix(sampling): keep stop sequence token ids aligned with their strings
#1456
opened Aug 10, 2026 by
DivyaNarahari97
Loading…
fix(pd): guard p_node before charging prefill load
#1454
opened Aug 10, 2026 by
DivyaNarahari97
Loading…
fix(api): harden OpenAI streaming/non-streaming error handling
#1426
opened Aug 4, 2026 by
sufubao
Collaborator
Loading…
perf(pd): reduce qwen3next prefill linear-att state memory
#1421
opened Aug 4, 2026 by
sufubao
Collaborator
Loading…
fix(multinode): use JSON instead of pickle for node-IP exchange to prevent RCE
#1419
opened Aug 2, 2026 by
Anai-Guo
Loading…
Previous Next
ProTip!
Updated in the last three days: updated:>2026-08-23.