Memvid
https://github.com/memvid/memvid
Single-file memory layer (one .mv2 file). No DB, no server. Append-only sequence of immutable Smart Frames with timestamps + checksums. Native Rust core (rewritten from Python).
At a glance
- Type
- Single-file append-only Smart Frames
- Tier
- T2
- Section
- Dedicated memory layers
- Created
- 2025-05
- Latest release
- v2.0.139 2026-03-13
- License
- Apache-2.0
- GitHub
- 15.3k★ +149/mo Rust
- Pricing
- OSS (Apache-2.0); free — no commercial pricing found
- Funding
- searched not found
Taxonomy
- storage
- file
- retrieval
- similarity
- persistence
- long-term
- update
- append-only
- unit
- chunk
- governance
- inspectable
- conflict
- append
When to use
Optimised for: simplicity / minimal config (one .mv2 file)
Anti-fit: not for multi-tenant / concurrent-write production workloads
Pros & cons
Pros
Single-file .mv2 store you can git commit — memory becomes a first-class artifact, not a database; sub-ms reads from a Rust core.
Cons
Single-file format means scale ceilings; non-standard format limits ecosystem interop.
Claims & capabilities
Sub-millisecond ops (0.025ms P50, 0.075ms P99). Reports +35% on LoCoMo. 10–100× speedup vs the prior Python implementation.
Technical surface
- API surface
- searched not found
- Backend storage
- searched not found
- Deployment
- Self-hosted only (single .mv2 file; no cloud service)
- Embedding model
- searched not found
- Multi-tenancy
- searched not found
- MCP
- native (first-party) — claude-brain MCP
- A2A
- not documented publicly
- OpenTelemetry
- not documented publicly
Similar systems
Other dedicated memory layers in the catalog, ranked by inbound references.
- Mem0 T1
Universal memory layer for AI agents. Three concurrent stores (vector + graph + KV); LLM-extracted facts; concurrent retrieval via ThreadPoolExecutor.
- Zep & Graphiti T1
Bi-temporal knowledge graph (event time + ingestion time). Strong on chronological reasoning and contradiction tracking. Graphiti is the open-source core.
- Cognee T1
"Extract–Cognify–Load" pipeline that turns raw input into a typed, queryable knowledge graph for agent recall.
- Hindsight (Vectorize) T1
Standalone memory service from Vectorize. Open source. Biomimetic four-network design (World, Bank, Observation, Opinion). Ships an MCP memory server.
- Supermemory T1
Memory engine with API, app, browser extension, and MCP server. Extracts facts, tracks updates, resolves contradictions, auto-forgets expired info. Plugins for Claude Code, OpenCode, OpenClaw, Hermes.
- AI Singapore SEA-LION T2
SEA-LION-Embedding (March 2026): retrieval + reranking models contrastively trained on 245M text pairs across 10 SE Asian languages. SEA-BED benchmark (169 datasets). SEA-LION v4 (Gemma-based) at 128K context with native function calling.
Related systems
Referenced by (2)
- claude-brain builds on — claude-brain packages memvid (same GitHub org memvid/) as a Claude Code plugin; cross-listings note flagged this pair.
- claude-brain same team as — Same GitHub organisation `memvid` publishes both repos.