Alex c88c188328 Local: raise LMCache shm_size 32→64 for LMCACHE_L1_GB=60 (124 GB rig) ...

Rig-specific tuning — this rig has 124 GB RAM (110 free), so L1=60
clears the preflight gate (L1+28=88) with headroom. shm_size must be
>= L1 or the MP connector falls back to slow pickle.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

2026-07-12 23:15:37 +02:00

..

agents-a1/vllm

fix(launch): runtime-agnostic GPU pinning via UUIDs — CDI/NixOS support (#610)

2026-07-06 22:59:51 +00:00

diffusiongemma-26b-a4b/vllm

fix(launch): runtime-agnostic GPU pinning via UUIDs — CDI/NixOS support (#610)

2026-07-06 22:59:51 +00:00

gemma-4-12b

Bump llama.cpp pin to server-cuda-b9967 across all composes (+4 think-ON) (#680)

2026-07-12 01:00:29 +05:00

gemma-4-26b-a4b

fix(launch): runtime-agnostic GPU pinning via UUIDs — CDI/NixOS support (#610)

2026-07-06 22:59:51 +00:00

gemma-4-31b

fix(launch): runtime-agnostic GPU pinning via UUIDs — CDI/NixOS support (#610)

2026-07-06 22:59:51 +00:00

ornith-1.0-9b/ik-llama/compose/single/deepreinforce-q4km

…

ornith-1.0-35b/ik-llama/compose/dual/deepreinforce-q8

…

qwen3-omni-30b-a3b/vllm-omni

…

qwen3.6-27b

Local: raise LMCache shm_size 32→64 for LMCACHE_L1_GB=60 (124 GB rig)

2026-07-12 23:15:37 +02:00

qwen3.6-35b-a3b

Promote vllm/qwen-35b-a3b-dual-nvfp4-fast: experimental -> caveats (#676)

2026-07-11 17:55:41 +05:00

qwen3.6-40b-deckard/llama-cpp/compose/dual/piehsoft-q6k

Bump llama.cpp pin to server-cuda-b9967 across all composes (+4 think-ON) (#680)

2026-07-12 01:00:29 +05:00

tess-4-27b

Record Tess vLLM A0 baseline: 106 off / 113 on — fallback triggered, slug stays 🧪

2026-07-12 05:48:39 +00:00

vibethinker-3b

Bump llama.cpp pin to server-cuda-b9967 across all composes (+4 think-ON) (#680)

2026-07-12 01:00:29 +05:00