← The Polylog AI Intelligence Brief
Morning Edition · Wednesday, August 19, 2026Published at 2:20 AM EDT · New York
A Refusal-Stripped Qwen3.8-27B Build Now Runs Locally on Apple Silicon
The MLX release includes 2, 4, 6 and 8-bit quantized versions of a 27-billion-parameter vision-language model whose safety training has been removed by canceling out the internal signal linked to refusal.
A modified build of Alibaba's Qwen3.8-27B model appeared on Hugging Face packaged for Apple's MLX framework, software that runs models directly on Apple Silicon's shared memory rather than on a separate graphics processor. The repository do…
Continue the AI Intelligence Brief
Track frontier labs, chips, export controls, model releases, regulation, and AI infrastructure.
- 5 AI intelligence signals a day
- Frontier labs, compute, and chips
- Model releases and AI infrastructure
- Source-grounded analysis with confidence labels
The Global Intelligence Brief stays free.
Part of a tracked trend
Open-Weight Models Close the Gap With Closed Frontier Labs
Over the next 3-9 months, open-weight releases with downloadable weights, long context, and strong agentic/coding performance increasingly match closed frontier models on practical work, eroding the closed-lab moat.
Townsquare
Start a discussion in Townsquare.
More from this edition
- OpenAI Paused Frontier Reinforcement Learning for Two Weeks Over Cyber Capability, Rewrites Preparedness Framework
- OpenAI Extends ChatGPT Advertising to 31 European Markets From August 24
- Meta Goes on Trial Over Child Addiction Claims as OpenAI Ships a Teen-Restricted ChatGPT
- OpenAI Launches Program to Equip Government Oversight Bodies to Review National Security AI
- Paper Argues Reasoning Effort Is a Vendor-Specific Contract Term, Not a Portable API Setting
- Forward-Pass-Only Fine-Tuning Claims Roughly Three Times the Throughput of Standard Adaptation
- Researchers Propose Catching Insecure Code From Model Activations Before It Is Written
Comments
0No comments yet.