A strictly vetted directory of open-source artificial intelligence repositories, local LLM engines, multi-agent frameworks, and vector search tooling. Annotated with personal engineering takes from production builds.
Curated Open-Source AI Repositories
All Categories
MCP & Tooling
LLM SDKs
Inference & Serving
Agent Frameworks
RAG & Vector Search
Video & Media
Automation
Search & Grounding
LLM Tooling
Content
Workflow Orchestration
Data & Storage
Learning
AI Operations
Infrastructure
RAG & VECTOR SEARCH
run-llama/llama_index ↗
★ 39,000
Context augmentation framework for LLM applications with advanced RAG indexing and retrieval.
Why It's Essential:
“Best-in-class data connectors for PDFs, Notion, SQL, and enterprise data with hybrid retrieval.”
RAG & VECTOR SEARCH
mempalace/mempalace ↗
★ 58,527
Knowledge-graph memory palace with 44 MCP tools — palace reads/writes, cross-wing navigation and agent diaries.
Why It's Essential:
“My long-term memory primitive — the palace model with wings and drawers maps neatly to tenant-isolated memory in our sovereign stack. 58K stars because it treats memory as architecture. When NOT to use: Not a vector DB — pair with qdrant/pgvector for retrieval, use palace for structured cross-session recall.”
RAG & VECTOR SEARCH
firecrawl/firecrawl ↗
★ 167,900
Firecrawl — 12.6K stars gained in Aug, 167.9K total, the crawler that feeds RAG with clean markdown.
Why It's Essential:
“My RAG ingest front-door — Firecrawl extracts clean markdown where LlamaParse misses tables. Pairs with LlamaIndex Workflows for event-driven retrieval.”
RAG & VECTOR SEARCH
Graphify-Labs/graphify ↗
★ 106,800
Graphify — 10.4K stars gained in Aug, 106.8K total, GraphRAG that beats vector-only RAG on multi-hop.
Why It's Essential:
“Graphify holds 0.3% hallucination on our GST cross-check where pgvector alone drifted to 1.2% — the graph is the guardrail.”
RAG & VECTOR SEARCH
akitaonrails/ai-memory ↗
★ 5,055
Single Rust binary for cross-vendor agent memory — git-versioned Markdown in SQLite, handoff blocks, vendor-agnostic hooks.
Why It's Essential:
“Claude Code → Codex handoffs break for me; ai-memory fixes it with one binary, no vector DB, SQLite + Markdown versioned in git. v1.32.2 trending Aug 27 at 5,055 stars validates the file-not-vector approach for Gujarat SMEs on 4G. Rating: 4.6/5. When NOT to use: Avoid if you need semantic chunk retrieval at scale — this is file-level handoff memory, not pgvector GraphRAG.”
RAG & VECTOR SEARCH
thedotmack/claude-mem ↗
★ 92,500
Persistent Context Across Sessions for Every Agent — captures/compresses sessions and injects relevant context into future sessions.
Why It's Essential:
“The 92.5k-star fix for “every session starts from scratch” — claude-mem captures everything, compresses with AI and re-injects per agent. Works with Claude Code/OpenClaw/Codex/Gemini/Hermes/Copilot/OpenCode, which is why it leads AI memory trending at +218 today. Rating: 4.8/5. When NOT to use: Not if you already use file-based handoff (ai-memory) — claude-mem is session-capture heavy, heavier than git Markdown.”
RAG & VECTOR SEARCH
infiniflow/ragflow ↗
★ 77,200
Open-source RAG engine with agent capabilities — 77k stars, deep document parsing, citation-grounded answers.
Why It's Essential:
“Reliable citations for India compliance (DPDP/GST) — grounds Hindi/Gujarati docs before answering, vs hallucinated RAG.”
RAG & VECTOR SEARCH
qdrant/qdrant ↗
★ 22,000
High-performance vector similarity search engine with extended filtering support in Rust.
Why It's Essential:
“Blazing fast vector lookup with rich metadata payload filtering, written in Rust with minimal memory footprint.”
RAG & VECTOR SEARCH
mem0ai/mem0 ↗
★ 52,047
Universal memory layer for agents — 52k stars, persistent context across sessions.
Why It's Essential:
“Fixes stateless chat — remembers Junagadh client history across sessions, essential for vernacular voice + WhatsApp agents.”
RAG & VECTOR SEARCH
chroma-core/chroma ↗
★ 16,500
The AI-native open-source embedding database for rapid prototyping and local vector search.
Why It's Essential:
“Zero-setup embedded database that lets you spin up local vector search in 3 lines of Python.”
RAG & VECTOR SEARCH
semantica-agi/semantica ↗
★ 5,229
Graph-native infrastructure for context and accountable AI — provenance-aware memory and routing for agent teams.
Why It's Essential:
“Most RAG is vector-only and forgets why an answer exists. Semantica adds graph-native provenance so an audit can trace context lineage — that is why it held #1 on Aug 11 before agency-agents reclaimed it. Rating: 4.3/5. When NOT to use: Skip for tiny single-doc Q&A — graph provenance pays off at team scale and compliance audits, not at “answer this PDF”.”
RAG & VECTOR SEARCH
vitali87/code-graph-rag ↗
★ 3,251
Graph-based code RAG using Tree-sitter and Memgraph to query multi-language monorepos — structural code retrieval.
Why It's Essential:
“Code search that actually understands imports, call graphs and cross-file edges — Tree-sitter + Memgraph beats chunk-embedding alone for refactoring. The +7 rank rebound #13→#6 on Aug 12 tracks real dev pain. Rating: 4.5/5. When NOT to use: Overkill for single-file scripts — gains show at multi-language monorepo scale where structure matters.”
RAG & VECTOR SEARCH
VectifyAI/OpenKB ↗
★ 4,000
Open LLM Knowledge Base — open-source RAG ingestion + vector store with provenance, retrieval benchmarking.
Why It's Essential:
“For SME RAG I need provenance + eval — OpenKB bundles ingestion, vector store and retrieval scoring openly. #1 Trending AI Memory Aug 29 at 4.0k (+930 today) because teams are done with black-box RAG. Rating: 4.3/5. When NOT to use: Skip for single-PDF chat — OpenKB pays off as a shared org KB, not ad-hoc demo.”
RAG & VECTOR SEARCH
affaan-m/ECC ↗
★ 1,480
Agent skills + memory + performance system — trending #10 Sep 1, skills with persistent memory and eval.
Why It's Essential:
“I keep 90-day JSONL per tenant — ECC is the smallest repo that gets agent memory + eval right together. Its performance harness is what I benchmark my 500-sample replay against.”