Commit Graph

10 Commits

Author SHA1 Message Date
keboss-m
80d665e948 Document deploy workflow in README.md and AGENTS.md 2026-06-10 14:27:35 +03:00
keboss-m
eee8f4c8a4 Replace LightRAG with native Python RAG engine + add deploy tooling
- New: src/rag/engine/ — in-process hybrid search (FTS5 BM25 + sqlite-vec + LLM rerank)
- New: src/rag/qmd/ — compatibility layer (qmd_query, qmd_chat, qmd_chat_stream, qmd_index_*)
- New: src/ingest/stub_writer.py — .md stubs for binary files (videos, archives)
- New: scripts/deploy.sh + scripts/pull_models.sh + Makefile + .env.example
- Removed: LightRAG, sentence-transformers embedding via separate package, rag_standalone/
- Removed: @nousresearch/qmd npm dep (package not published); Node.js from Dockerfile
- Updated: tests/ (46 passed), docker-compose, .dockerignore, config.yaml, README

Engine: in-process Python (no daemon, no npm), sentence-transformers 384-dim,
RRF fusion (k=60), BM25 + vector with numpy fallback. WebSocket API unchanged.

Deploy: 'git clone' + 'make init' + 'make pull-models MODELS_SOURCE=...' + 'make up'.
Models (5.83 GB) live outside git; pulled via rsync from dev host.
2026-06-10 14:24:01 +03:00
Кирилл Блинов
22eb20a2db Clarify: large-v3 is max, bad audio needs preprocessing not bigger models 2026-05-29 18:42:35 +03:00
Кирилл Блинов
6e5ee64be0 Move HF_TOKEN to .env file for one-command docker compose up 2026-05-29 18:11:00 +03:00
Кирилл Блинов
0ed45cdf12 Integrate HF_TOKEN into Docker build for preloaded diarization models 2026-05-29 18:04:38 +03:00
Кирилл Блинов
b9897555a3 Update README with web service documentation 2026-05-29 12:17:27 +03:00
Кирилл Блинов
c771f83351 Add multi-format output support (docx + md simultaneously) 2026-05-29 11:39:13 +03:00
Кирилл Блинов
bd7eadb49f Update README with video support and first-run documentation 2026-05-29 10:53:44 +03:00
Кирилл Блинов
de212f5f00 Add local/offline processing explanation to README 2026-05-29 10:29:51 +03:00
Кирилл Блинов
bdd94b860f Add README and fix .gitignore for markdown files 2026-05-29 10:28:16 +03:00