Haiku RAG vs Visual Memory MCP
Two MCP servers from our AI & ML ranking, compared on live data - updated nightly, never sponsored.
Bottom line · 2026-09-18
Visual Memory MCP records significantly more monthly package installs (21,590 vs 4,907); both are actively maintained, with commits inside the last month.
How they differ in kind
Haiku RAG focuses on: Agentic RAG that answers questions about your own documents with citations to page numbers and section headings. Visual Memory MCP focuses on: @putervision/vision-memory-mcp is a zero-infrastructure, local-first Model Context Protocol (MCP) server and CLI tool that provides AI coding assistants (such as Cursor, Claude Code, Gemini, or… Implementation languages differ (Python vs TypeScript).
What each one does
Haiku RAG
Agentic RAG that answers questions about your own documents with citations to page numbers and section headings. Runs locally on an embedded database, no server required.
From the project's README.
Visual Memory MCP
@putervision/vision-memory-mcp is a zero-infrastructure, local-first Model Context Protocol (MCP) server and CLI tool that provides AI coding assistants (such as Cursor, Claude Code, Gemini, or Copilot) with visual state caching using perceptual hashing, local CLIP embeddings, and transition graphs to eliminate repetitive vision LLM calls.
From the project's README.
14 signals, side by side
| Signal | Haiku RAG | Visual Memory MCP |
|---|---|---|
| Monthly installs | 4,907 | 21,590 |
| GitHub stars | 609 | 70 |
| Last commit | 2026-09-17 | 2026-09-15 |
| Releases · last 90 days | 10+ | 3 |
| In the registry since | Oct 2025 | Jul 2026 |
| Registry versions | 132 | 11 |
| Documented tools | not extracted - see README | not extracted - see README |
| Env vars extracted | none extracted - see README | none extracted - see README |
| Runs | local (stdio) | local (stdio) |
| Endpoint auth | local only | local only |
| Maintenance | actively maintained | actively maintained |
| License | MIT | NOASSERTION |
| Language | Python | TypeScript |
| Category rank | #4 of 106 | #3 of 106 |
Which one, for what
Derived from the signals above, not hands-on testing.
Pick Haiku RAG if…
- → release velocity matters - 10+ releases in 90 days vs 3
- → clear licensing matters - MIT vs undeclared
Pick Visual Memory MCP if…
- → you want the more widely installed option - 21,590 monthly installs vs 4,907
- → you prefer a TypeScript codebase to extend or audit
Who's behind them
Haiku RAG - ggozad: 1 MCP server tracked, 1 maintained, 609 combined stars.
Visual Memory MCP - putervision: 6 MCP servers tracked, 6 maintained, 210 combined stars. Full record.
Not sold on either?
The next-ranked AI & ML servers we track:
Quick answers
Is Haiku RAG better than Visual Memory MCP?
Package installs favor Visual Memory MCP: 21,590 monthly installs to 4,907. Haiku RAG still ranks #4 of 106 maintained AI & ML servers. Data as of 2026-09-18; we haven't hands-on tested either.
Can I use Haiku RAG and Visual Memory MCP together?
Yes - MCP clients accept multiple servers in one config, so you can enable both AI & ML servers side by side. If their tools overlap, keep the one whose toolset fits to keep your agent's tool list lean.
Keep exploring: best AI & ML servers · Python servers · TypeScript servers · all comparisons · every server
Methodology: package-usage signal = npm/PyPI installs (last month, platform APIs) · maintenance = commit recency + release cadence (GitHub) · tool lists and environment-variable names extracted from each project's README · endpoint auth from our own nightly probes. We haven't hand-tested these servers; everything here is data as of 2026-09-18.