#1313th of 18 in Memory
memvid
Single-file memory store for AI agents, with search built in
Live demo ↗
(opens in a new tab)Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 16.6k
- License
- Apache-2.0
- Last commit
- Jul 2026
- Last release
- May 2026
- Language
- Rust
Missed the latest GitHub check; these facts may be older.
Overview
Memvid packs documents, embeddings, a Tantivy full-text index, an HNSW vector index and a time index into one append-only .mv2 file, so no database server is needed. It is a Rust library (memvid-core) with a CLI and Node.js and Python SDKs. Optional features cover local ONNX text embeddings, OpenAI embeddings, CLIP image search, Whisper transcription and encrypted capsules.
Who it is for: Developers adding embedded long-term memory to agents without a vector DB
Strengths
- One .mv2 file holds data and indexes; no .wal, .lock or sidecar files
- Append-only frames with embedded WAL give crash recovery and queries over past states
- Runs offline with local ONNX embeddings (BGE, Nomic, GTE) and Whisper models
- SDKs for Rust, Node.js and Python plus a CLI; Apache-2.0 licensed
Weaknesses
- Local embedding ONNX models must be downloaded manually with curl before use
- Benchmark claims (+35% on LoCoMo, 0.025ms P50) come from the project itself
- README gives no multi-user or concurrent-writer details for the single file
- Features like predictive caching and branching are not shown in README examples
What it needs
- GPU optional
- Needs Rust 1.85.0+, ONNX embedding model files (for vec), OpenAI API key (for api_embed only)
- Models: bge-small-en-v1.5, bge-base-en-v1.5, nomic-embed-text-v1.5, gte-large, OpenAI text-embedding-3
Also in Memory
See all 18| Rank | Project | Score |
|---|---|---|
| 1 | gbrainLong-term memory store for coding agents, with sources and corrections | 85 out of 100 |
| 2 | hindsightAgent memory server with retain, recall and reflect operations | 80 out of 100 |
| 3 | CogneeMemory engine that turns documents and code into a knowledge graph | 79 out of 100 |
| #4 | MemPalaceLocal long-term memory store for agents with verbatim storage and semantic search | 78 out of 100 |
| #5 | agentmemoryPersistent cross-session memory for coding agents via hooks, MCP and REST | 77 out of 100 |