Hemory

Hemory is an AI memory tool that captures conversations to create searchable memory for AI agents.

Hemory captures conversations through always-on or scheduled listening, turning audio into a private, searchable memory for AI agents via Model Context Protocol (MCP). It stores raw audio locally on your device with zero cloud audio retention, and uses voice-activity detection to filter silence and background noise. It supports mainstream agents and clients like Claude Code, Codex, Gemini, Cursor, VS Code, OpenClaw, and Hermes. You can use it to draft documents, build work report decks, write nightly journals, and generate product requirement documents from customer interviews. It offers manual and scheduled listening modes, 100-language transcription, and speaker recognition by voice.

Hemory at a glance

Best for: Researcher, Software engineer
Runs on: macOS, Windows, Linux, Web, iOS
Pricing: Free Trial $0 one-time; Starter $10/mo; Pro $20/mo; Max $50/mo
Released: September 2026
Category: AI research tool
Website: hemory.com

Key features

Capture always-on listening
Connect agents via MCP
Search private memory
Record audio locally
Filter silence with voice-activity detection
Transcribe 100 languages
Recognize speakers by voice

Use it for

Build monthly work report decks from recorded conversations
Write nightly journals summarizing daily events
Draft product requirement documents from customer interviews
Answer questions about past meetings and discussions

Alternatives to Hemory

past.dev — Provides agents with accurate, timestamped context instead of static retrieval.
Memori — Reduces LLM token usage and spend by over 97% with targeted recall.
Turnbook — Saves every chat turn locally with zero model tokens and shares memory across AI apps.
DocuMemory — Turns messy AI chat transcripts into clean, reusable context for future sessions.
Deep Recall — Keep your personal context and history when switching between different AI tools.
Hy Memory — Provides specialized memory pipelines for chat, working, and coding tasks.
chat-recall — Search every conversation and track AI findings in one place.
IAI Memory Engine — Provides persistent, local, and encrypted memory for AI agents without APIs.
Playground — Persists agent memory across sessions and searches by meaning.
HippoRAG — Improves multi-hop retrieval and complex sense-making without sacrificing simple task performance.
M-Flow — It achieves a 93.7 percent top-one hit rate using a conical memory architecture.
memsearch — It shares continuous context across Claude Code, Codex, and other AI coding agents.

All Hemory alternatives