haylan Arthur Erlich
0 Followers · 0 Following
Schweiz
https://me.arthurerlich.de
Joined on 2025-05-30
Repositories
16
 Projects Packages Public Activity Starred Repositories
1
haylan pushed to main at haylan/LLM-Server 2026-09-02 21:53:45 +00:00
124053cf89 feat(scripts): auto-register memory-and-notes in litellm's DB for the Admin UI
haylan pushed to main at haylan/LLM-Server 2026-09-02 21:35:27 +00:00
47bdb22457 docs: record issue #24's smoke-test findings and local litellm-pgvector patches
7e8b5069d4 fix(scripts): update.sh never applied litellm-pgvector's db schema
eaf11b6d4b fix(scripts): ingest-memory.sh never actually computed embeddings
213550e44b fix(litellm): prefix EMBEDDING__MODEL with openai/ for litellm-pgvector
3eda4e3ec0 fix(litellm-pgvector): honor caller-supplied id on vector store create
Compare 7 commits »
haylan pushed to main at haylan/LLM-Server 2026-09-02 21:08:01 +00:00
e2dd106f74 feat(scripts): fold model downloads into update.sh, skip if already present
haylan pushed to main at haylan/LLM-Server 2026-09-02 21:00:00 +00:00
fb7cfc9148 fix(litellm): raise qwen3.8-27b-local max_tokens floor 4096 -> 16384
haylan pushed to main at haylan/LLM-Server 2026-09-02 20:55:45 +00:00
30a8523433 fix(lazytainer): drop network_mode: host, join ai-stack (fixes self-ID panic)
haylan pushed to main at haylan/LLM-Server 2026-09-02 20:48:02 +00:00
c7848f356a chore(scripts): mark update.sh executable
haylan pushed to main at haylan/LLM-Server 2026-09-02 20:47:20 +00:00
cb9b3a9045 fix(docs): update stale MEMORY_RETRIEVAL_EMBEDDING_KEY refs after litellm-pgvector revert
e7983f0710 Revert "feat(knowledgebase): replace litellm-pgvector connector with memory-retrieval"
Compare 2 commits »
haylan pushed to main at haylan/LLM-Server 2026-09-02 20:28:33 +00:00
b4dc83949e feat(litellm): add Redis for router state/rate-limits/budgets/cache
haylan pushed to main at haylan/LLM-Server 2026-09-02 20:21:57 +00:00
b996b1fe88 fix(scripts): update.sh now appends vars missing from .env, not just blank ones
haylan pushed to main at haylan/LLM-Server 2026-09-02 20:18:18 +00:00
24d749b2e0 feat(scripts): fold generate-secrets.sh into update.sh, auto-mint virtual keys
haylan commented on issue haylan/LLM-Server#24 2026-09-02 20:10:03 +00:00
Verify search + knowledgebase wiring (#21) on the real R9700 box

Updated scope after #25: this ticket now verifies memory-retrieval (the langchain_postgres.PGVector-based service, services/memory-retrieval/) instead of the retired litellm-pgvector…

haylan closed issue haylan/LLM-Server#25 2026-09-02 20:09:50 +00:00
LangChain+pgvector direct RAG vs. the litellm-pgvector connector — which fits this stack?
haylan commented on issue haylan/LLM-Server#25 2026-09-02 20:09:49 +00:00
LangChain+pgvector direct RAG vs. the litellm-pgvector connector — which fits this stack?

Implemented — switched from litellm-pgvector to a LangChain-direct service (services/memory-retrieval/), per this ticket's verdict. Details in the commit on main (abeadc4): new `memory-retriev…

haylan pushed to main at haylan/LLM-Server 2026-09-02 20:09:39 +00:00
abeadc49c8 feat(knowledgebase): replace litellm-pgvector connector with memory-retrieval
e2a79eab4a docs(research): LangChain+pgvector-direct vs litellm-pgvector for issue #25
Compare 2 commits »
haylan commented on issue haylan/LLM-Server#25 2026-09-02 20:03:18 +00:00
LangChain+pgvector direct RAG vs. the litellm-pgvector connector — which fits this stack?

Verdict: switch to LangChain-direct. Full research + a concrete replacement-service spec: docs/research/langchain-pgvector-vs-litellm-pgvector.md on branch `research/langchain-pgvector-vs-lit…

e2a79eab4a docs(research): LangChain+pgvector-direct vs litellm-pgvector for issue #25
haylan created branch research/langchain-pgvector-vs-litellm-pgvector in haylan/LLM-Server 2026-09-02 20:03:16 +00:00
haylan opened issue haylan/LLM-Server#25 2026-09-02 19:55:46 +00:00
LangChain+pgvector direct RAG vs. the litellm-pgvector connector — which fits this stack?
haylan pushed to main at haylan/LLM-Server 2026-09-02 19:50:52 +00:00
7247674a9f fix(litellm-pgvector): vendor the source instead of a remote git build context
haylan pushed to main at haylan/LLM-Server 2026-09-02 19:44:33 +00:00
b0e37b2f4c chore(scripts): add update.sh, hook it into agent/deploy workflow