What it is
Memory-native LLMs that persist and reuse latent state across sessions without replaying full context. Includes SGLang serving and snapshot-based memory restore. Serve chat models that store past interactions in latent native memory, reuse them across turns and sessions, and answer from memory without keeping the original text in context.
Key results
- Persistent state is ~66× smaller than full KV cache at 128K history
- MemT-1.0-Flash and Nano rank 1st and 2nd on 11 memory benchmarks
- Supports memory snapshots and restore via saved server-side state files
Shows that long-horizon conversational memory can be carried in fixed-size latent state instead of replaying full text history.