Article Brief Key Takeaways 4 Points24s Read 01Agent tools–The browser-side JavaScript sandbox can now activate llama.cpp’s agentic request flow even without a server-advertised tool. 02Daily workflow–Conversation bulk actions and settings expand the web UI’s application…
Fenghe Open Weights Lack Forecasting Proof
July 18, 2026
China’s weather agency has made Fenghe available to outside developers. The immediate proof is unusually concrete: model weights, configuration files and a tokenizer are public. The harder proof is still missing. Nothing in the release…
Article Brief Key Takeaways 4 Points24s Read 01Control surface–CrewAI 1.15.3 adds interception points around agent execution boundaries, including model and tool operations. 02Safer default–Tool-result caching is now opt-in, making data reuse a deliberate application decision….
Article Brief Key Takeaways 4 Points24s Read 01Control surface–custom_remat lets AI framework code express forward, rematerialization and backward behavior at a function boundary. 02Opt-in path–The new behavior depends on experimental remat3, whose configuration flag remains…
Inkling’s Open Weights Still Need a Data Center
July 15, 2026
Article Brief What matters 4 Points24s Read 01Scale–Only 41 billion of Inkling’s 975 billion parameters are active per token, but the smallest official self-hosting floor is still 600 GB of VRAM. 02Customization–Tinker offers managed LoRA…
llama.cpp's 4.26× Intel Gain Has a Narrow Catch
July 15, 2026
The 4.26× number is the part of llama.cpp b10016 that will travel fastest. It is also the part most likely to lose its denominator. The result comes from a contributor-run test on one Intel Arc…