MemT-1.0: memory foundation models with native persistent memory
MemTensor releases MemT-1.0-Nano (36B total, 3B active) and MemT-1.0-Flash (288B, 17B active), open models that build persistent memory into the model itself instead of a retrieval layer.
Why it matters
Constant-size memory state: about 66x less persistent storage and about 1.86x faster inference than full-context KV cache at 128K history. It ranks first and second among evaluated parametric-memory methods, though strong text-retrieval memory systems still lead overall on detail-heavy long histories.
What you could build with it
Run MemT-1.0-Nano as the memory backbone of a long-running coding agent and compare task success against a RAG-memory baseline on your own repo history.
Built with MemT-1.0
- MemTensor/MemOS repositoriestheresanaiforthat · Repository listing for MemTensor's MemOS with cloud and local OpenClaw plugins, docs, and dashboard for persistent agent memory.
- MemT-1.0 Memory-Native LLMs cut KV storage 66x at 128K tokens (AI Weekly)aiweekly · Deep read of the paper: 74 authors, Qwen3.6-35B-A3B and DeepSeek-V4-Flash backbones, 7.6-point SWE-bench Verified gain from internalized procedural memory.
- MemTensor/MemOSgithub · 11,965 ★ · MemTensor's full memory OS — cloud and local plugins for OpenClaw/Hermes agents with hybrid retrieval and tiered skill evolution; the practical packaging of their persistent-memory research.
First spotted on huggingface: source.
More AI models
OpenAI publishes 722 AI-generated math manuscripts from an unreleased modelOpenAI released 722 mathematical manuscripts in 372 result families to GitHub under Apache-2.0, produced by an…models · JEV 0.74Tencent Hunyuan Hy3: open 295B MoE with 21B active under Apache 2.0Tencent open-sourced Hy3, a 295B-parameter mixture-of-experts model with only 21B parameters active per token, under…models · JEV 0.73NVIDIA Isaac Lab 3.0 GA + Isaac OS open runtime + Agile ONE SN humanoidMajor Isaac platform expansion at GTC Berlin Oct 21: Isaac Lab 3.0 is generally available as the open-source robot…models · JEV 0.73StepFun Step-5-Preview — open BF16 weights releasedStepFun published the full BF16 weights for its flagship Step-5-Preview agentic model (600B-parameter sparse MoE, 27B…models · JEV 0.73NVIDIA Nemotron 3.5 Retriever collection: Embed 8B/1B, Rerank 8B, Parse 631MNVIDIA's Oct 21 collection drop for agentic retrieval: Nemotron-3.5-Embed in 8B BF16, 1B BF16 and 1B NVFP4…models · JEV 0.72Kimi K3.1 Pro: Moonshot's 2.8T open-weight flagship officially launchesMoonshot AI officially launched Kimi K3.1 Pro on Oct 23, a mixture-of-experts model with 2.8 trillion total / 104…models · JEV 0.72
Get the week's best AI launches, plus 3 ideas worth building
One email every Saturday. Ranked by traction, not hype. Free.