Self-evolving memory OS for LLM & AI Agents: ultra-persistent memory, hybrid-retrieval, and cross-task skill reuse, with 35.24% token savings
MemTensor/MemOS is drawing steady momentum at +16.6 stars/day (79th percentile in the tracked cohort), for a momentum score of 40/100. A recent release is helping fuel the run.
Reconstructed from 261 snapshots — the time-series GitHub’s API doesn’t expose.
Release count rose to 60.
~67 forks/day, 9.1× the trailing baseline.
~27 forks/day, 7.2× the trailing baseline.
Live score, refreshed every 6 hours. Links back to this page.
[](https://repoclimb.com/repo/MemTensor/MemOS)Low breakout odds over the next 14 days, led by star velocity (79th pctl) and push recency (97th pctl). Confidence is high given the available history.
Transparent heuristic · logged for model training
Percentile rank within the tracked cohort. The score self-calibrates — it answers “accelerating vs. everything else,” not raw size.
~10 forks/day, 7.7× the trailing baseline.
Release count rose to 59.
Release count rose to 58.
Release count rose to 57.
Release count rose to 56.
~10 forks/day, 5.9× the trailing baseline.
Release count rose to 55.
Release count rose to 54.
Release count rose to 53.