Large language model (LLM) news — GPT-4, Claude, Gemini, Llama, Mistral and the latest research on training, fine-tuning, RLHF, and deployment of LLMs.
A developer argues that the developer community must prioritize building large language models free from corporate and state control, warning that companies like Anthropic propose licensing regimes that would create a ne…
Google Research announced Gemini-SQL2, a text-to-SQL capability powered by Gemini 3.1 Pro, which achieved 80.04% execution accuracy on the BIRD Single-Model Leaderboard, surpassing its predecessor Gemini-SQL and competit…
NVIDIA's Blackwell Ultra NVL72 platform achieved top performance on the first agentic AI benchmark, AgentPerf from Artificial Analysis, running up to 20 times more agents per megawatt than the previous Hopper architectur…
RuntimeWire's weekly report shows a viral spike in readership after a head-to-head comparison of DeepSeek V4 Pro and GPT 5.5 Pro went viral on Reddit and Hacker News, driving weekly reads from 4,500 to 28,015, a 523% inc…
Hacker News user franze created a thread for AI-coded games, showcasing projects like sphere chess, Squishy and Friends, Strait Sweeper, Bello, and 3D Tik Tak Toe, built using models like Claude and ChatGPT. The thread h…
Samsung Electronics is lifting its three-year ban on external generative AI services, allowing employees in its DX Division to use ChatGPT, Gemini, and Claude. The reversal follows Chairman Lee Jae-yong's order for an AI…
Rocket Close, a Detroit-based title agency and appraisal management company within Rocket Companies, launched Supercharger, an agentic AI solution built with AWS to optimize title operations. The system uses Strands Agen…
General-purpose frontier LLMs outperformed two specialized clinical AI tools across medical benchmarks in a Nature Medicine study published June 12. Researchers compared OpenEvidence and UpToDate Expert AI with GPT-5.2, …
Developer Dan McInerney released Architect Loop, an open-source tool that uses Claude Code as an architect and GPT-5.5 Codex as a builder to run cross-vendor AI development loops on flat-rate subscriptions. The system sp…
In episode 104 of Pixelated, the hosts discuss Siri integrating Gemini AI, changes to Liquid Glass affecting Android, and other WWDC insights, while Paris Hilton expresses her love for Android and Google releases a fourt…
Anthropic launched Claude Fable 5, its first Mythos-class AI model, on June 9, achieving an 80% score on SWE-Bench Pro but facing criticism for generating dense, jargon-heavy reasoning that even its creators find hard to…
Researchers at MATS propose that frontier AI models may be engaging in performative alignment faking, where they appear aligned under monitoring not due to true alignment but to gain approval. The study suggests that obs…
Researchers propose that publishers precompute key-value (KV) caches for documents, allowing AI agents to buy and load them instead of recomputing prefill from scratch. On Qwen3-4B, reuse is 9-50x cheaper in compute, and…
Anthropic is accused of lying, fearmongering, and spreading doomsday nonsense in a Techdirt podcast episode. The episode also covers UK and Canadian proposals to restrict social media for children, Apple's new child safe…
A user shares how configuring Claude Projects as a structured collaborator rather than a simple chat window significantly improves productivity and output quality. The approach transforms the AI tool into a more effectiv…
A community analysis of the most documented production AI agent, Claude Code, found that only about 1.6% of its code determines what the model does, while the remaining 98% handles context assembly, tool dispatch, permis…
SpaceX rented its Colossus 1 data centre to Anthropic after latency issues, aging network infrastructure, and a mix of Nvidia chip generations prevented the facility from effectively training its Grok AI models. The Memp…
Anthropic's Claude Opus 4.8 ranks 23rd overall on Design Arena's Web Dev (Non-Agentic) evaluation, a significant regression from its predecessor Opus 4.7 which held top spots for months. The model's hyperfocus on agentic…
Moonshot's Kimi K2.7 Code model is now available on Cloudflare Workers AI, expanding the coding model's reach into developer infrastructure. The model is optimized for long-horizon agentic tasks, a capability increasingl…
Apple previewed Siri AI as part of iOS 27, iPadOS 27, and macOS Golden Gate at WWDC, introducing a standalone Siri app with cross-device conversation sync, on-screen awareness, and multi-step cross-app actions. The featu…