Arthur Erlich

haylan merged pull request haylan/LLM-Server#51

2026-09-07 17:56:27 +00:00

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

haylan pushed to feat-rag-databases at haylan/LLM-Server

2026-09-07 17:55:08 +00:00

7a654ead91 chore: ignore .qwen/temp

haylan pushed to feat-rag-databases at haylan/LLM-Server

2026-09-07 17:40:59 +00:00

2bfe6dbd29 docs: point knowledge.proxy-ai.home at Neo4j's browser

haylan created pull request haylan/LLM-Server#51

2026-09-07 17:35:39 +00:00

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

haylan pushed to remove-fast-model at haylan/LLM-Server

2026-09-07 17:33:57 +00:00

5d6a17fd9b feat: remove llama-server-fast (Qwen3-4B classifier model)

haylan created branch remove-fast-model in haylan/LLM-Server

2026-09-07 17:33:57 +00:00

haylan created pull request haylan/LLM-Server#50

2026-09-07 17:24:16 +00:00

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

haylan pushed to feat-rag-databases at haylan/LLM-Server

2026-09-07 17:20:26 +00:00

20f2ec3ab2 feat: add qdrant and neo4j for RAG vector/graph storage

haylan created branch feat-rag-databases in haylan/LLM-Server

2026-09-07 17:20:26 +00:00

haylan commented on issue haylan/LLM-Server#44

2026-09-06 20:34:29 +00:00

Add llama-server-fast: small non-thinking classifier/fast model for qwen-code Auto Mode

Confirmed working end-to-end on the real server: llama-server-fast deployed, registered in OmniRoute (qwen3-4b-fast), wired into qwen-code's fastModel, and the Auto Mode classifier now works…

haylan closed issue haylan/LLM-Server#5

2026-09-06 20:30:07 +00:00

Verify the stack on the real Radeon R9700 box (GPU passthrough + Lazytainer idle-stop)

haylan commented on issue haylan/LLM-Server#5

2026-09-06 20:30:06 +00:00

Verify the stack on the real Radeon R9700 box (GPU passthrough + Lazytainer idle-stop)

Confirmed: GPU_MAX_HW_QUEUES=1 + the numeric-GID group_add fix (#47) resolved the 100%-GPU-pin with both llama-server and llama-server-fast up. Combined with the earlier classifier-context and…

haylan commented on issue haylan/LLM-Server#5

2026-09-06 20:27:21 +00:00

Verify the stack on the real Radeon R9700 box (GPU passthrough + Lazytainer idle-stop)

Confirmed on the real server (#49, merged): the interactive gum-based config sync in scripts/update.sh works as designed — the conflict screen appeared for the stale LLAMA_FAST_PARALLEL=2 vs…

haylan deleted branch fix-fastmodel-context-size from haylan/LLM-Server

2026-09-06 20:21:54 +00:00

haylan pushed to main at haylan/LLM-Server

2026-09-06 20:21:54 +00:00

5b7548dc7c Merge pull request 'Fix llama-server-fast context-size exhaustion breaking Auto Mode' (#49) from fix-fastmodel-context-size into main

ea7b05fb99 feat: interactive per-key conflict resolution in update.sh's config sync

9def240a8e feat: update.sh force-syncs tracked config from .env.example

52a92f6508 fix: llama-server-fast context-size exhaustion breaking Auto Mode classifier

Compare 4 commits »

haylan merged pull request haylan/LLM-Server#49

2026-09-06 20:21:53 +00:00

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

haylan pushed to fix-fastmodel-context-size at haylan/LLM-Server

2026-09-06 20:20:31 +00:00

ea7b05fb99 feat: interactive per-key conflict resolution in update.sh's config sync

haylan pushed to fix-fastmodel-context-size at haylan/LLM-Server

2026-09-06 20:10:54 +00:00

9def240a8e feat: update.sh force-syncs tracked config from .env.example

haylan created pull request haylan/LLM-Server#49

2026-09-06 20:07:17 +00:00

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

haylan pushed to fix-fastmodel-context-size at haylan/LLM-Server

2026-09-06 20:07:02 +00:00

52a92f6508 fix: llama-server-fast context-size exhaustion breaking Auto Mode classifier