Modelsslug: qwen-3-8

Qwen 3.8

A version of Alibaba's Qwen model family, deployed locally in the demo to auto-build robot simulation environments.

Curated Intelligence

1 items · Page 1/1
量子位 (微信公众号)T2·Infra
62 pts
Nvidia RTX Spark Mini PC Packs 128GB Unified Memory for Local 120B Models

Nvidia unveiled the RTX Spark mini PC and a laptop variant built on the same chip, which fuses CPU and GPU into a single package with 128GB of unified memory, enough to run 120B-parameter models and large Unreal Engine 5 scenes locally, starting at $4,399. In demos, the device drove an end-to-end robot training workflow, using a locally deployed Qwen model to auto-build simulation environments and control the open-source microduck robot.