PrismML has released Ternary Bonsai, a family of 8B, 4B, and 1.7B LLMs using 1.58-bit ternary weights.
Qwen3.6 is the latest Qwen model family with strengthened agentic coding and thinking-context preservation.
Agents SDK and sandbox safely automate legacy code migration.
Frontier models are harder to prevent **distillation** of than expected, and large-scale pretraining can easily collapse from small **bias** and **causality** violations.
Nvidia is solidifying its AI infrastructure dominance through supply chain, financial, and China strategies.
JavaScript was disabled on X, so the article body could not be loaded.
Vercel Workflows moves to GA, handling long-running agents and backends inside your code.
Apple will unveil numerous poster, oral, and workshop presentations at ICLR 2026.
Anthropic has launched Claude Opus 4.7, a flagship model with self-verification capabilities and high-resolution vision.
Google's Gemma 4 model supports on-device AI inference that runs on iPhone without an internet connection.
Computron autonomously performs web browsing, security checks, and bug fixes every day.
A case report that unused Haiku 4.5 and Gemini 3 Flash requests were counted toward billing on Copilot Pro+.
Alibaba's Qwen team has released Qwen3.6-35B-A3B, an open-source MoE-based model with significantly improved agentic coding performance.
The Claude Opus 4.7 model card specifies synthetic data generated by other models.
Comparing Vulkan and ROCm on a Strix Halo 128GB machine showed the winner varying by model.
California's AG has raised allegations of Amazon's price control practices based on unsealed documents.
An AI agent automates everything from microscope mapping to safe pin probing.
Qwen3.6 has been added to the Ollama library and can be run right away.
A case of running Qwen 3.6 35B on an RTX 2000 Ada 16GB at 400 t/s prompt and 24 t/s generation.
Claude Opus 4.7 beat Claude Opus 4.6 by 1 keystroke in Vim Golf.
For Somali TTS, XTTS V4 works best so far, but the LLM and Whisper layers remain bottlenecks.
Tree-sitter has greatly improved editing, search, and linting in the R ecosystem.
Released a Skill and verification harness that helps port transformers to mlx-lm.