abiba-bot/inference-harness
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
Files
54a4f26db7230a99712766aeeb51f2d714fa9fb4
inference-harness/router
T
History
Abiba 54a4f26db7 fix: Default tier back to Dense-first (MoE overheating at 91°C) 
Heavy tier keeps MoE primary (workhorse for >25K tok).
Default tier routes Dense → VLM → MoE to prevent MoE overload.
MoE had 5 timeouts in 15 min when Default pushed overflow to it.
2026-05-28 21:40:18 +00:00
..
Dockerfile
May 19, 2026: Full harness update
2026-05-19 15:03:47 +00:00
http_patch.py
May 19, 2026: Full harness update
2026-05-19 15:03:47 +00:00
requirements.txt
May 19, 2026: Full harness update
2026-05-19 15:03:47 +00:00
router.py
fix: Default tier back to Dense-first (MoE overheating at 91°C)
2026-05-28 21:40:18 +00:00
router.py.bak.20260518074236
May 19, 2026: Full harness update
2026-05-19 15:03:47 +00:00
ts_patch.py
May 19, 2026: Full harness update
2026-05-19 15:03:47 +00:00