omega-memory/omega-memory

MCP Developer Tools

Local‑first, cross‑model memory that lets AI agents retain context and learn across sessions without sending data to the cloud

Track releases GitHub Website

Python Latest v1.4.9 · 1mo ago Security brief →

Features

Stores decisions, preferences, and error patterns with semantic search for recall across agent sessions
Multi‑agent coordination tools (file locking, task queues, intent broadcasting) to prevent conflicts
Intelligent LLM routing that selects the optimal model per task based on cost and capability
Knowledge base ingestion of PDFs, markdown, web pages, and files with semantic chunking

Recent releases

View all 41 releases →

v1.4.9 New feature 1mo

Notable features

Reranker precision selection via OMEGA_RERANKER_PRECISION=int8 or download_model(precision="int8")

Full changelog

Features

Reranker precision selection -- Set OMEGA_RERANKER_PRECISION=int8 to use the quantized model (~571 MB on disk, ~650 MB RSS) instead of the FP32 default (~2.3 GB on disk, ~2.5 GB RSS). Also available via download_model(precision="int8"). Auto-detect finds whichever variant is installed.

Upgrade

pip install --upgrade omega-memory

View release on GitHub

v1.4.8 Bug fix 1mo

Security fixes

Depersonalized codebase and hardened credential handling

Full changelog

Fixes

fix: download_model() 404 on bge-reranker-v2-m3 — BAAI upstream repo no longer ships ONNX weights. Repointed to onnx-community/bge-reranker-v2-m3-ONNX (HuggingFace's canonical ONNX conversion). Also added model.onnx_data sidecar and fixed download verification logic. Closes #53.
security: depersonalize codebase and harden credential handling

Upgrade

pip install --upgrade omega-memory

View release on GitHub

v1.4.7 Bug fix 1mo

⚠ Upgrade required

Run `pip install --upgrade omega-memory` then execute `omega activate OMEGA-PRO-` to enable Pro features.

Full changelog

Fix: Pro activation works from bare omega-memory install

omega activate now works without requiring the Pro wheel to be pre-installed:

Validates the license key directly against the activation API
Auto-downloads and installs the Pro package
Single command from install to fully working Pro

Previously, omega activate failed with "OMEGA Pro is not installed" because it required a package that only exists after Pro setup, creating a chicken-and-egg problem.

Upgrade

pip install --upgrade omega-memory
omega activate OMEGA-PRO-<your-key>

View release on GitHub

v1.4.6 Feature 1mo

Notable features

Free‑tier protocol fallback with pro upgrade reminder

Changelog

Free-tier protocol fallback with pro-reminder

View release on GitHub

v1.4.5 New feature 1mo

Notable features

Pro‑reminder system with graduated urgency based on memory count
GitHub star ask added to CLI status card, setup, and welcome briefing

Full changelog

What's Changed

Added

Pro-reminder system with graduated urgency based on memory count
GitHub star ask in CLI status card, setup, and welcome briefing
Tool gate: upgrade CTA when free users call any of 85 Pro-only tools (was bare "Unknown tool" error)

Changed

Periodic reminder interval reduced (less frequent, more contextual)
Milestone warnings now range-based (1,000+/1,800+) instead of exact counts
Session stop hook shows memory count vs limit at graduated frequency
All LLM-facing upgrade notices use "Tell the user:" instruction format so agents relay to humans
Degradation warnings start earlier with approaching-limit context