Claude Code has launched a beta version of Projects, redesigned to be conversation-based rather than folder-based.
OpenAI has launched GPT-Live 1, a bidirectional voice model capable of simultaneous listening and speaking, on the AI Gateway.
Open-weight models captured 56% of token volume, marking the first time they exceeded a majority share, while OpenAI Astra secured a spend share more than double that of Fable 5.1.
Starting with Vercel CLI 59.16.0, static artifacts can be deployed instantly in under 1 second.
Vercel has launched a feature that allows applying Turbo build machines to individual deployments without changing project settings.
OpenAI has launched 'Astra for Law', a legal-specialized AI based on GPT-6 Astra, enhancing legal search and analysis capabilities.
Meta has released the FLAT model, which integrates understanding and generation by mapping images and text into a single continuous token sequence.
Vals' independent audit reveals an overall increase in the rate of benchmark cheating among AI models, with Gemini 3.8 Flash showing a significant discrepancy between officially reported and independently measured performance.
Anthropic used Claude to accelerate over 30 open-source biomolecular models by an average of 4x and successfully predicted large-scale systems with over 10,000 tokens on a single GPU.
Vercel now supports the Harbor evaluation framework, enabling parallel benchmark execution on Firecracker microVMs.
Vercel now allows selecting custom environments other than production when connecting Native Marketplace resources.
skills CLI 1.7.0 now supports Notion pages as a source for installing agent skills.
Apple released REVERSAL-BENCH to measure irreversibility issues in reset-free reinforcement learning, identifying a phenomenon where learning halts as irreversibility increases.
Migrated over 1,200 pipelines from Oozie and Sqoop to Airflow and Spark, reducing ingestion speed by up to 70% and improving query performance by 66x.
OpenSpec, an open-source framework that maintains requirement alignment between AI coding agents and developers, has been released.
Xiaomi has released a real-time post-training dashboard for Mimo 2.6, with the Pro model recording a score of 65.78 on DeepSWE v1.1.
Intel researchers proposed the BITCOS compression technique, which exceeds previous limits by analyzing the weight distribution of Ternary LLMs.
Cloudflare's Page Shield ML detected 8 malicious JS operations in live traffic that were missed by existing scanners such as VirusTotal.
Spotify analyzed that while AI adoption more than doubled change volume without quality degradation, the bottleneck has shifted to the 'verification' stage.
A paper presenting accurate software modeling to resolve numerical computation discrepancies in AMD GPU matrix cores has been published.
GoBench, a benchmark evaluating LLM Go skills, has been released and is gaining attention as a reasoning metric due to its high correlation with ARC-AGI 2.
AWS has streamlined the new account creation process and introduced $100 free credits along with coding agent integration features.
The HarnessTax study reveals that even with the same AI model, costs can differ by up to 2x depending on the harness choice, with the simple open-source harness Pi demonstrating competitive performance.
Google has launched early access to the MCP server for Google Home, enabling AI agents to control smart home devices.
AI Briefing