MemT-1.0: Towards Memory-Native Evolution
To this end, we introduce MemT-1.0, a new architecture of memory foundation models that internalize memory as part of model computation.
ProofPaper ↗
Key points
- Memory-native evolution requires making memory an intrinsic capability of foundation models, enabling them to continuously accumulate and utilize information.
- Building on Metis, MemT-1.0 employs Gated DeltaNet-2 with a stable token-level memory update mechanism within each turn.
- We further introduce Mixture-of-Storage to scale native memory capacity.
- We release a model family comprising MemT-1.0-Nano and MemT-1.0-Flash, and conduct a comprehensive evaluation of memory and general capabilities.
Sources (2)
- [1]MemT-1.0: Towards Memory-Native EvolutionHugging Face Daily Papers · Oct 19, 12:00 AM
To this end, we introduce MemT-1.0, a new architecture of memory foundation models that internalize memory as part of model computation.
Memory-native evolution requires making memory an intrinsic capability of foundation models, enabling them to continuously accumulate and utilize information.
- [2]MemT-1.0: Towards Memory-Native EvolutionarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 19, 05:54 PM · same content
Extractive summary: sentences quoted from the sources.
Before this
- Oct 18, 2026Google Open-Sources Harness2, a Recursive Agent Harnessing Framework
- Oct 17, 2026deepseek-ai/DeepSeek-V4.1-Flash-1017
- Oct 17, 2026Nemotron 3.5 Bolt, Muse Spark Open, and a Fast LFM2.5 24B
- Oct 16, 2026Improved Qwen3.8-Flash Chat Template (w/ Sharp Template terse option)
- Oct 16, 2026Qwen Flash Next potential SOTA Q2 nobody talks about
- Oct 16, 2026Welcome Beam: a 501B open model you can own, customize, and scale