Recall – local multimodal semantic search for your files
patel_aayushya
17 points
8 comments
April 06, 2026
Related Discussions
Found 5 related stories in 677.1ms across 14,015 title embeddings via pgvector HNSW
- Show HN: Recall – Local project memory for Claude Code mateenah · 95 pts · June 21, 2026 · 60% similar
- OpenAI reinvents Recall except everything is stored remotely pluc · 12 pts · April 22, 2026 · 55% similar
- Show HN: Adaptive Recall, persistent memory for AI assistants over MCP abratabia · 20 pts · July 12, 2026 · 53% similar
- Memanto; open-source memory agent that remembers, recalls and answers majidfekri · 14 pts · June 18, 2026 · 49% similar
- Show HN: Antfly: Distributed, Multimodal Search and Memory and Graphs in Go kingcauchy · 85 pts · March 17, 2026 · 49% similar
Discussion Highlights (4 comments)
patel_aayushya
Search "team dinner" and surface the photo — even with zero metadata. Powered by Gemini Embedding 2 + local ChromaDB. Comes with an animated setup wizard and a Raycast extension for instant visual search. Nothing leaves your machine except embedding API calls.
esafak
Does it consider relationships between files, esp. for code? Otherwise I don't see how it would work.
TZubiri
Great, let's just hope aayu22809 ( whose github activity started in yesterday, and whose commits are coauthored by claude) doesn't compromise the supply chain of the company.
ajsnigrutin
Everything you have on your pc is sent to AI... "local", yeah. If you want local but text-only search, there is a very similarly named project called Recoll - https://recoll.org/ - a bit ugly, but works great, especially for people who tend to save notes in many different text files in many different folders.