// memory api · v0.2.0 · uptime 99.97%

Memory for AI apps. Built like infrastructure, not lifestyle.

Persistent user memory in two endpoints. Synthesized profiles drop straight into a system prompt — no chunks to deduplicate, no ORM to maintain, no embedding pipeline to babysit.

latency p50
47 ms
latency p95
198 ms
embedding
768-dim
retrieval
BM25 + vector
POST /v1/ingest · GET /v1/context200 OK · 47ms
// POST /v1/ingest await memory.ingest({ userId: "user_abc", content: "Prefers TypeScript. Building a VAPI voice agent.", }); // → 202 Accepted { id, queued: true } // GET /v1/context const ctx = await memory.context({ userId: "user_abc" }); // → 200 OK // { // static: ["Senior TS engineer", "Uses BullMQ", ...], // dynamic: ["Building VAPI voice agent this week"], // relevant: [...] // }
ingest p50
23ms
▼ 4ms · 7d
context p50
47ms
▼ 2ms · 7d
cache hit %
62%
▲ 5pt · 7d
connectors
4
notion · gdocs · linear · transcripts
retention
pro
7d free · ∞ pro
uptime · 30d
99.97%
no incidents

Two endpoints. No SDK required.

SDKs are convenience wrappers. The contract is REST, the auth is a bearer token, the response is JSON. Two calls do the job; the other two surfaces are there when you need them.

POST/v1/ingest

Store a conversation turn, document, note, or meeting transcript. Sanitization + chunking + embedding happen async — request returns in <30ms.

GET/v1/context

Returns synthesized static + dynamic arrays plus optional relevant chunks. Drop both into a system prompt.

POST/v1/search

Raw hybrid search (vector + BM25 + RRF). For when you want the chunks, not the profile.

GET/v1/entities

Temporal entity graph extracted during synthesis. People, projects, decisions, with valid-time edges. Pro+.

Synthesized profiles. Not raw chunks.

Every other memory tool returns vector chunks you still have to deduplicate, rank, and trim. Anansi runs a synthesis pass on every ingest and returns a versioned profile that drops directly into your system prompt.

01 · Synthesis, not search

A Claude pass distills accumulated chunks into static facts (stable truths) and dynamic context (current state). Drop both arrays into a system prompt.

02 · Hybrid retrieval

Vector + BM25 + reciprocal rank fusion under the hood. Catches semantic matches AND exact lookups. Knobs exposed when you need them.

03 · 4 connectors built-in

Notion, Google Docs, Linear, meeting transcripts. OAuth in 2 clicks. Sync runs in the background. The synthesis layer doesn't care where the chunks came from.

04 · Model-agnostic

Claude, GPT, Llama, Gemini, on-prem. You pick the brain; we're the memory. Bring your own embedding vectors if you've already invested in a pipeline.

Where Anansi sits.

An honest read of the memory landscape. Updated when the field shifts.

capabilityanansisupermemorymem0zep
Synthesized profile (system-prompt ready)
Hybrid retrieval (BM25 + vector)vec onlyvec only
Temporal entity graph
Built-in connectors48+
Hosted SaaS (no infra to run)OSSself-host
Free tier1k ingest/molimitedOSS

Pay for what you ingest. No per-seat tax.

Usage-metered. Cancel any time via /portal. Free tier is real — 1,000 ingest calls per month, no card.

free
$0/mo
  • 1,000 ingest calls
  • 500 context calls
  • 10 memory users
  • 7-day retention
  • vector retrieval
Start free
scale
$99/mo
  • 250,000 ingest calls
  • 100,000 context calls
  • everything in pro
  • priority support
  • SLA on request
Contact us

Anansi didn't weave a new web for every story. Your agent shouldn't build a new mind for every session.

Stop letting your agent forget.

Two function calls. Any LLM. Free tier, no card. Five minutes from signed-up to remembering users in production.