kyleandClaude Fable 5.1 09d16f2f39 llama-server: cap ornith 9B helper output (runaway guard) ...

As Hermes' auxiliary model it received requests with no max_tokens and a
generation ran to 46k tokens (~70 min) after the client's 60 s timeout,
starving the 35B. Cap thinking (reasoning-budget 2048) and total output
(n-predict 4096) server-side; preset verified against a throwaway router.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

2026-09-21 01:13:30 -07:00

..

imladris

add ono-sendai

2025-12-02 14:03:47 -08:00

ono-sendai

add ono-sendai

2025-12-02 14:03:47 -08:00

orion

Cleanups

2026-07-08 02:42:38 -07:00

rift

Cleanups

2026-07-08 02:42:38 -07:00

sk

add sk

2025-09-09 16:21:14 -07:00

straylight

llama-server: cap ornith 9B helper output (runaway guard)

2026-09-21 01:13:30 -07:00

svc

Add svc host: NixOS config for TornadoVPS edge node

2026-04-04 15:21:01 -07:00

vade

Cleanups

2026-07-08 02:42:38 -07:00