grenade

grenade merged pull request helexa/helexa#28

2026-06-12 11:15:39 +00:00

feat: keep auto-recovering models visible as recovering (#20)

grenade closed issue helexa/helexa#20

2026-06-12 11:15:39 +00:00

neuron: keep auto-recovering models listed as recovering in /models (so cortex holds the route)

grenade pushed to main at helexa/helexa

2026-06-12 11:15:39 +00:00

02f20bc9e1 Merge pull request 'feat: keep auto-recovering models visible as recovering (#20)' (#28) from feat/neuron-20-recovering-status into main

2a231e49de merge main (sccache enablement supersedes branch cuda-check pin)

9bb9678f93 fix(ci): pin CUDA_COMPUTE_CAP in cuda-check — builder image has no nvidia-smi

df9c490614 feat(neuron+gateway): keep auto-recovering models visible as recovering (#20)

Compare 4 commits »

grenade deleted branch feat/neuron-20-recovering-status from helexa/helexa

2026-06-12 11:15:39 +00:00

grenade pushed to feat/neuron-20-recovering-status at helexa/helexa

2026-06-12 11:06:00 +00:00

2a231e49de merge main (sccache enablement supersedes branch cuda-check pin)

2dadea5d8d ci: enable sccache on the build jobs (conditional on the CUDA image)

Compare 2 commits »

grenade pushed to main at helexa/helexa

2026-06-12 11:05:31 +00:00

2dadea5d8d ci: enable sccache on the build jobs (conditional on the CUDA image)

grenade pushed to feat/neuron-20-recovering-status at helexa/helexa

2026-06-12 10:55:27 +00:00

9bb9678f93 fix(ci): pin CUDA_COMPUTE_CAP in cuda-check — builder image has no nvidia-smi

grenade pushed to main at gongfoo/gongfoo

2026-06-12 10:51:42 +00:00

a609f27b55 fix(agent): per-runner SCCACHE_SERVER_PORT to stop cross-container server sharing

grenade created pull request helexa/helexa#28

2026-06-12 10:42:20 +00:00

feat: keep auto-recovering models visible as recovering (#20)

grenade created branch feat/neuron-20-recovering-status in helexa/helexa

2026-06-12 10:42:07 +00:00

grenade pushed to feat/neuron-20-recovering-status at helexa/helexa

2026-06-12 10:42:07 +00:00

df9c490614 feat(neuron+gateway): keep auto-recovering models visible as recovering (#20)

grenade pushed to main at helexa/helexa

2026-06-12 10:24:07 +00:00

f5fa840dfb ci: escalate sccache retries — restart server, then fall back uncached

grenade pushed to main at gongfoo/gongfoo

2026-06-12 10:19:18 +00:00

f838d0f9ef fix(brew): budget scale-up against available runners, not busy ones

grenade commented on issue helexa/helexa#27

2026-06-12 10:18:41 +00:00

tracking: prioritised path to closing every open issue

Re-prioritisation note (per the working agreement here): iteration latency jumped the queue ahead of P1, by operator decision — the ~20.5 min push-to-testable tax (measured on the 2026-06-08…

grenade pushed to main at helexa/helexa

2026-06-12 10:17:27 +00:00

7557c5e877 ci: cut iteration latency — change-aware builds, gated deploys, dev fast path

grenade pushed to main at gongfoo/gongfoo

2026-06-12 10:05:25 +00:00

d909e9a86e feat(agent): capacity caps for shared-purpose hosts; add benjy + quadbrat as cuda build hosts

grenade pushed to main at gongfoo/gongfoo

2026-06-12 09:53:04 +00:00

b45e76bb04 feat(reaper): reap zombie runners whose Gitea task was cancelled

grenade pushed to main at gongfoo/gongfoo

2026-06-12 09:17:08 +00:00

9a314425ea fix(images): install file(1) in all base runner images

grenade opened issue helexa/helexa#27

2026-06-12 09:02:07 +00:00

tracking: prioritised path to closing every open issue

grenade commented on issue helexa/helexa#15

2026-06-12 08:58:14 +00:00

Vision: numerical validation against transformers reference

Posture note (2026-06-12 reframing): this issue's priority rose. The README now publicly claims every served architecture is "implemented in this repository, ported against the HuggingFace…