Launching Tenor: Open-Source 4D Memory Architecture for AI Agents
We just launched Tenor, an open-source 4D spatio-temporal memory architecture for AI agents.
It replaces static vector search with continuous timeframes, signed Hebbian associative plasticity, and sub-350-token bounded prompt quotas. The system runs completely locally without requiring external database infrastructure.
Install and run locally:
git clone https://github.com/exorticlabs/tenor.git cd tenor && pip install -e . Quick links:
- GitHub: https://github.com/exorticlabs/tenor
- Hugging Face: https://huggingface.co/exorticlabs/tenor
- Research Paper: https://github.com/exorticlabs/tenor/blob/main/docs/RESEARCH_PAPER.md
- Website: https://exortic.com