Skip to main content

Pricing Overhaul — 2026-02-15

Changed

  • Free endpoints: List, get, delete, search (text), suggested, core memories, relations, history, graph, export, namespaces, and stats are now completely free — no payment required
  • Paid endpoints now only include those using OpenAI (embeddings or LLM):
    • Embedding endpoints: store (0.005),store/batch(0.005), store/batch (0.04), recall (0.005),update(0.005), update (0.005)
    • LLM + embedding endpoints: extract (0.01),consolidate(0.01), consolidate (0.01), ingest (0.01),context(0.01), context (0.01), migrate ($0.01)
  • Free tier reduced from 1,000 to 100 calls (only paid endpoints consume credits)

v1.0.0 — 2026-02-13

Added

  • Hybrid retrieval: 4-signal scoring — vector similarity + keyword (BM25) + recency + importance weighting
  • Connection-weighted decay: Memories with more relations decay slower
  • Context-dependent importance: Importance dynamically boosted by query relevance
  • Adaptive keyword boost: Strong keyword matches (>0.3) get higher weight
  • Memory relationships: related_to, derived_from, contradicts, supersedes, supports
  • Zero-effort ingestion (/v1/ingest): Dump a conversation, get extracted facts with dedup and auto-relations
  • LLM-based fact extraction (/v1/memories/extract): Extract structured facts from conversations
  • Memory consolidation (/v1/memories/consolidate): Rule-based and LLM-powered merging
  • Proactive suggestions (/v1/suggested): Surface stale, fresh, hot, and decaying memories
  • Memory pinning: Exempt memories from type-based decay
  • Memory types: correction, preference, decision, project, observation, general — each with configurable half-lives
  • Batch store: Up to 100 memories per request
  • Deduplication: Automatic on store (cosine similarity ≥ 0.95)
  • Free tier: 100 API calls per wallet with wallet signature auth
  • x402 micropayments: USDC on Base via EIP-3009 and Permit2
  • Dashboard: Wallet-authenticated web UI with memory browser, graph view, suggestions, and analytics
  • Session auth: Sign once with wallet, get JWT for subsequent requests
  • MCP server: Full integration with Claude Desktop, Cursor, and other MCP clients
  • Python SDK: Sync and async clients with typed models
  • TypeScript SDK: Full API coverage with TypeScript types
  • CLI: Command-line tool for all operations

Fixed

  • Soft-deleted memories excluded from all queries
  • Expired memories excluded from all queries