Radar / AI infrastructure / Tencent Hy Memory 2.0
Tencent Hy Memory 2.0 — production-grade dual-system cognitive memory engine
Tencent Hunyuan open-sourced Hy Memory 2.0 (PyPI hy-memory 2.0.1), the Hy-Memory monorepo, the open-weight Hy-Memory-30B-A3B model, and MemHub — a local-first desktop/CLI product that unifies conversation history and coding traces across Claude Code, Cursor, Codex, OpenCode, and Qwen Code into one managed memory runtime.
Why it matters
First major open-source memory stack from a Chinese frontier lab combining framework + specialized memory-writing model + cross-app desktop product, with claims of #1 on 4 memory benchmarks.
What you could build with it
A startup building long-running coding assistants could plug a dual-system memory engine into its agent so that project conventions, user preferences, and prior bug fixes persist across sessions instead of being re-learned in every conversation.
Does it hold up?
Too early to judge the 2.0 release itself; the lineage goes back to Hy Memory 1.0 (May 2026) and an EMNLP 2026 paper, and it ships both an SDK and a desktop product — but no independent third-party deployment reports for 2.0 exist yet.
Built with Tencent Hy Memory 2.0
- tencent/Hy-Memory-30B-A3B on Hugging Facehuggingface · Open-weight 30B-A3B model specialized in long-term memory writing, with vLLM quickstart.
- Technical report: attributed reinforcement learning for long-term memory writingarxiv · Companion paper behind the open-sourced memory model.
- Tencent-Hunyuan/Hy-Memorygithub · Official monorepo: hy-memory Python SDK, dual-system memory pipelines (chat/work/code), MemHub desktop+CLI, docs and demo video.
Learn more
First spotted on github: source.
More AI infrastructure
DeepGEMM: DeepSeek's open GPU BLAS kernel libraryDeepSeek's DeepGEMM, a clean and efficient open-source BLAS kernel library for NVIDIA GPUs (up to 1550 TFLOPS on…infra · JEV 0.77OpenViking — self-evolving context database for AI agentsByteDance Volcengine's open-source (AGPL-3.0) context database that unifies agent memory, knowledge RAG, and skills…infra · JEV 0.76Modular open-sources MAX inference server, Mojo stdlib and accelerator kernelsThe core of Modular's unified AI deployment stack is now open: the MAX inference server (OpenAI-compatible endpoints)…infra · JEV 0.76OmniRouteFree MIT AI gateway: one endpoint, 290+ providers and 500+ models with quota-aware auto-fallback.infra · JEV 0.75Context Mode: context-window optimization for AI coding agentsMCP server for AI coding agents that sandboxes verbose tool output and persists session memory, cutting context use by…infra · JEV 0.74StrataOpen-source inference engine that runs the 125B-parameter Qwen3.8-Flash-Next MoE on a single consumer GPU with 12GB+…infra · JEV 0.74
Get the week's best AI launches, plus 3 ideas worth building
One email every Saturday. Ranked by traction, not hype. Free.