mirror of
https://github.com/simstudioai/sim.git
synced 2026-09-24 15:45:35 +08:00
Waleed
ed23330f88
* feat(knowledge): hybrid lexical + vector retrieval for KB search KB search ranked purely on pgvector cosine distance, which retrieves exact tokens (error codes, ticket keys, identifiers, rare product names) poorly. Add a full-text leg over the already-present generated `embedding.content_tsv` column and its GIN index — no migration, no re-indexing — and fuse it with the vector leg by reciprocal rank. Both legs run concurrently and share the same visibility and tag-filter predicates; the lexical leg is best-effort and falls back to vector-only on failure. Hybrid is the default for every caller. `searchMode: 'vector'` on the internal and v1 contracts (and an advanced Retrieval Mode dropdown on the Knowledge block) restores the previous behavior. Both search routes now share one `executeKnowledgeSearch` dispatch instead of duplicating the three-branch retrieval logic. * change(knowledge): make vector the default search mode, hybrid opt-in Every existing caller — workflow block, v1 API, copilot, guardrail RAG — keeps its current ranking. Hybrid retrieval is now requested explicitly via `searchMode: 'hybrid'`. Also routes the copilot knowledge tool through the shared `executeKnowledgeSearch` dispatch so all four callers share one retrieval path, and documents `searchMode` on the public v1 search endpoint in the OpenAPI spec. * docs(knowledge): document the hybrid retrieval mode Regenerates the knowledge integration reference for the new searchMode tool param, and adds a Retrieval Mode section to the knowledge base workflow guide explaining when hybrid beats vector-only. * fix(knowledge): stop rank fusion from starving the lexical leg Rank n in one leg always ties rank n in the other, so ordering the fused list by score alone let whichever leg was scored first take every tied slot. At topK=1 that meant a hybrid search returned exactly the vector-only result and discarded the exact keyword match the mode exists to recover. Selection now orders by score and drains each tie group round-robin, taking from whichever leg has contributed fewest rows so far. The lexical leg is passed first so it wins a total tie, since a chunk the vector leg ranked below its distance threshold is the case hybrid was opted into for. * fix(knowledge): credit a shared hit to every leg that returned it Attributing a row found by both legs to a single leg left the round-robin owing the other leg a slot it had already been served. With a shared rank-1 hit and topK 2, that evicted the lexical-only row — the exact match hybrid was enabled to recover — in favor of the vector-only one. A shared row satisfied every leg that returned it, so every one of them is now charged for it. Tie-breaking prefers the candidate whose least-served leg has been served least, which also removes the arbitrary best-rank attribution. * fix(knowledge): reject a whitespace-only copilot query explicitly The shared dispatch treats a whitespace-only query as absent and throws when no tag filters accompany it, where the previous vector-only call would have embedded the blank string and searched. Tighten the existing guard so the tool returns its normal message instead. * fix(knowledge): fan the keyword leg out per knowledge base The vector leg caps candidates per base once getQueryStrategy sets useParallel, but the keyword leg always ran one global query with a single LIMIT. Searching several bases at once let whichever one ranks strongest lexically consume every slot, so an exact-token hit in a smaller base never reached fusion — the case hybrid exists to serve. The keyword leg now uses the same strategy: per-base queries under the same parallel limit, re-ranked globally on a selected ts_rank_cd. Both legs draw candidates the same way, so fusion combines rankings over the same pool. * perf(knowledge): stop the keyword leg detoasting every match's vector Selecting the cosine distance in the ranking query made Postgres detoast the 1536-dimension embedding and compute a distance for every full-text match before the LIMIT applied, so cost tracked how common the query term was rather than topK. On a 20k-chunk base with a term matching every row that was 61,055 buffer hits against 1,030 for the same query without the projection. Rank on ids and ts_rank_cd alone, then hydrate only the rows that survive the limit. Same results, and the worst case drops to ~27ms end to end.
A workspace to build, deploy and manage AI agents and workflows.
Quickstart
Cloud-hosted: sim.ai
Self-hosted
git clone https://github.com/simstudioai/sim.git && cd sim
bun run setup
Capabilities
- Connect 1,000+ integrations and every major LLM
- Add Slack, Notion, HubSpot, Salesforce, databases, and more
- Build agents visually, conversationally, or with code
- Ingest files, knowledge bases, and structured table data
- Monitor runs, logs, schedules, and workflow activity
One workspace, every surface
Chat and workflows are just the start — tables, files, knowledge, and scheduled tasks all live in the same workspace.
Tables — a database, built in |
Files — one store for your team and every agent |
Knowledge — your agents' memory |
Scheduled tasks — runs on your schedule |
Self-hosting
bun run setup is an interactive wizard: it provisions the database, generates secrets, writes your .env files, connects a Chat API key, and starts Sim the way you choose:
- Local dev — run from source to contribute or hack on Sim
- Docker Compose — a self-contained instance for testing self-hosting
- Kubernetes (Helm) — deploy to a local cluster
When it finishes, open http://localhost:3000.
Manage your install with bun run sim:
bun run sim start | stop | restart # bring your install up / down / cycle
bun run sim status # what's installed and healthy
bun run sim logs # follow logs
bun run sim doctor # diagnose configuration problems
bun run sim down # remove containers (data kept)
bun run sim reset # archive .env and wipe managed data
sim detects how you're running (Docker Compose, local dev, or Kubernetes) and acts accordingly.
Prefer a bare sim? Run bun link once — but note sim lands in ~/.bun/bin, which Homebrew's bun doesn't add to your PATH, so you may need export PATH="$HOME/.bun/bin:$PATH" in your shell profile.
Sim also supports local models via Ollama and vLLM. See the self-hosting docs for details.
Chat API Keys
Chat is a Sim-managed service. bun run setup connects a Chat API key for you — sign in when it opens your browser and the key is stored automatically. To view, create, or revoke keys later, go to sim.ai/selfhost/settings/chat-keys.
Environment Variables
See the environment variables reference for the full list, or apps/sim/.env.example for defaults.
Tech Stack
Next.js · Bun · PostgreSQL · Drizzle · Better Auth · Tailwind — and the rest of the stack
- Framework: Next.js (App Router)
- Runtime: Bun
- Database: PostgreSQL with Drizzle ORM
- Authentication: Better Auth
- Schema Validation: Zod
- UI: Shadcn, Tailwind CSS
- Streaming Markdown: Streamdown
- State Management: Zustand, TanStack Query
- Flow Editor: ReactFlow
- Docs: Fumadocs
- Monorepo: Turborepo
- Realtime: Socket.io
- Background Jobs: Trigger.dev
- Remote Code Execution: E2B
- Isolated Code Execution: isolated-vm
Contributing
We welcome contributions! Please see our Contributing Guide for details.
License
This project is licensed under the Apache License 2.0 - see the LICENSE file for details.
