-
Notifications
You must be signed in to change notification settings - Fork 0
Pull requests: ddvnguyen/hydra_vortex
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
docs(469): live-GPU parity smoke evidence — RPC PREFILL checkpoint honesty at fork HEAD 45c1435da
#754
opened Sep 9, 2026 by
ddvnguyen
Owner
Loading…
test(baseline): #740 baseline-concurrency — arms 115-129c, spike probes, arm111 pin
#749
opened Sep 8, 2026 by
ddvnguyen
Owner
Loading…
2 of 3 tasks
chore(submodule): llama-cpp → d06537485 baseline — #747 --parallel-ctx-threshold gate
#748
opened Sep 8, 2026 by
ddvnguyen
Owner
Loading…
fix(baseline): prime measured prompt/slots by default in concurrent-decode-test.sh
#745
opened Sep 7, 2026 by
ddvnguyen
Owner
Loading…
3 tasks done
test(baseline): #703 baseline-concurrency arms 092-102
#741
opened Sep 6, 2026 by
ddvnguyen
Owner
Loading…
fix(core): #738 — debounce capability clearing: require 2 consecutive empty-cap INFO polls
bug
Something isn't working
core
Core coordinator/scheduler logic
#739
opened Sep 3, 2026 by
ddvnguyen
Owner
Loading…
test(cold-atomic): failing repro for self-lease decode deadlock + fix design
#711
opened Aug 29, 2026 by
ddvnguyen
Owner
Loading…
epic/697-final-verify: integrate #695 (V2 goldens re-baselined) + minifleet evidence
#707
opened Aug 28, 2026 by
ddvnguyen
Owner
Loading…
v2 worker scheduler rewrite — SOLID state machine + differential parity (Closes #591)
status:review
PR open, agent review in progress
#695
opened Aug 21, 2026 by
hydra-z
Bot
Loading…
fix(core): prevent slot leak when client disconnects during T3 rebuild
#694
opened Aug 20, 2026 by
ddvnguyen
Owner
Loading…
feat(engine): worktree-proof llama.cpp build cache + build tuning
#685
opened Aug 16, 2026 by
ddvnguyen
Owner
Loading…
docs: fix stale Architecture and COMBINED engine mode sections in CLAUDE.md
#585
opened Aug 8, 2026 by
ddvnguyen
Owner
Loading…
3 of 4 tasks
ProTip!
Type g p on any issue or pull request to go back to the pull request listing page.