LLM-Server

docs(research): add qwen3.8-27b decode speed analysis

#68

by haylan was merged 2026-11-11 15:52:21 +00:00

main

docs/decode-speed-research

Feat/dual gpu

#67

by haylan was merged 2026-11-04 06:23:17 +00:00

main

feat/dual-gpu

Docs cleanup: restructure, move code comments into docs, remove image generation

#65

by haylan was merged 2026-09-24 20:01:30 +00:00

main

docs/cleanup-and-restructure

4 / 4

Fix/omniroute pr agent timeout

#54

by haylan was merged 2026-09-09 14:29:13 +00:00

main

fix/omniroute-pr-agent-timeout

fix(update.sh): stop config sync loop from dying silently on a missing key

#53

by haylan was merged 2026-09-07 18:10:23 +00:00

main

feat-rag-databases

Add Qdrant/Neo4j RAG storage + update.sh gum fixes

#52

by haylan was merged 2026-09-07 18:07:48 +00:00

main

feat-rag-databases

feat: remove llama-server-fast (Qwen3-4B classifier model)

#51

by haylan was merged 2026-09-07 17:56:27 +00:00

main

remove-fast-model

feat: add qdrant and neo4j for RAG vector/graph storage

#50

by haylan was merged 2026-09-07 17:56:49 +00:00

main

feat-rag-databases

Fix llama-server-fast context-size exhaustion breaking Auto Mode

#49

by haylan was merged 2026-09-06 20:21:53 +00:00

main

fix-fastmodel-context-size

Run git pull first in update.sh, not mid-script

#48

by haylan was merged 2026-09-06 19:40:10 +00:00

main

fix-update-sh-pull-order

Fix GPU pinned at 100% with two containers, flaky render group

#47

by haylan was merged 2026-09-06 19:35:32 +00:00

main

fix-gpu-pin-and-render-group

Downloader for Qwen-Image weights, switch-model.sh

#46

by haylan was merged 2026-09-06 19:01:02 +00:00

main

comfyui-model-and-switch-script

Add llama-server-fast: small non-thinking classifier/fast model

#45

by haylan was merged 2026-09-06 18:39:31 +00:00

main

add-fast-model

feat(llama.cpp): raise default context to 128K, document RAM/SSD offload knobs

#30

by haylan was merged 2026-09-03 04:44:34 +00:00

main

ctx-size-128k

fix(litellm): default max_tokens=4096 for the reasoning model

#20

by haylan was merged 2026-09-02 18:58:12 +00:00

main

fix/litellm-reasoning-max-tokens

feat(litellm): add UI_USERNAME/UI_PASSWORD for the admin UI login

#19

by haylan was merged 2026-09-02 18:29:30 +00:00

main

fix/litellm-ui-admin-creds

fix(downloader): run as root to fix permission denied on models volume

#18

by haylan was merged 2026-09-02 18:03:59 +00:00

main

fix/downloader-permission-denied