Large language model (LLM) news — GPT-4, Claude, Gemini, Llama, Mistral and the latest research on training, fine-tuning, RLHF, and deployment of LLMs.
New research from Flinders University reveals that harsh criticism from video game fans is a form of brand loyalty, not rejection. Analyzing over 23,000 Reddit comments using AI, the study found that Call of Duty and Bat…
An AI research engineer has open-sourced their entire workflow and prompts for coding agents, releasing a public workflow playbook on GitHub. The engineer critiques the high cost and overthinking tendencies of the Fable …
Antigravity, an agentic AI coding assistant, autonomously engineered a Rust-based agentic interface that allows users to control OBS Studio via natural language. The tool features a reasoning loop powered by Claude and O…
A developer outlines principles for building production AI applications on Cloudflare using Claude Code, emphasizing simplicity, treating AI models as engineering teams, and leveraging Cloudflare's edge stack including W…
MarkTechPost published a tutorial on building memory-efficient Transformers using xFormers, covering packed sequences, grouped-query attention, ALiBi, SwiGLU, and causal attention. The guide demonstrates how to achieve f…
A developer built AETHER, a fully offline, voice-controlled AI assistant that runs three local language models on a laptop to control a PC, send WhatsApp messages, and perform desktop automation without cloud APIs. The s…
Haystack released a tutorial on building a generative question-answering pipeline using retrieval-augmentation (RAG). The pipeline integrates SentenceTransformersTextEmbedder, InMemoryEmbeddingRetriever, ChatPromptBuilde…
An experiment comparing Kimi K2.7 Code and Claude Fable 5 across 12 landing pages found that Kimi cost 94% less while achieving similar quality scores. The open-source model was roughly 16 times cheaper than Fable and 8 …
Hermes Agent by NousResearch surpassed 153,000 GitHub stars in under three months, becoming a fast-growing AI developer tool. The deciding factor for developers choosing between Hermes Agent and Claude Code plus LoreConv…
A developer details their daily AI coding stack, using both an editor-integrated AI tool and a chat model for different tasks, and notes that AI excels at boilerplate and test scaffolding but is unreliable for security-s…
Developers are increasingly switching to local LLMs for coding tasks, with two Hacker News threads (1,260 and 1,350 points) confirming local models are now viable. The guide compares self-hosted inference (Ollama, llama.…
GitHub Copilot engineers Ryan Caldwell and Bhavya U detailed recent improvements to token efficiency in the agentic harness, including extended prompt caching, tool search, and WebSocket connections, to reduce costs and …
OpenRouter launched Fusion, a routing layer that sends a single prompt to multiple AI models in parallel and synthesizes their outputs, achieving performance comparable to Anthropic's Claude Fable 5 at roughly half the c…
AI coding tools are reshaping software engineering, shifting focus from writing code to evaluating what to build and judging output quality. Anthropic's Mythos model found thousands of bugs but patched few, highlighting …
Anthropic's Model Context Protocol (MCP) enables standardized context integration for LLMs. Google has published a guide for deploying a remote MCP server on Google Kubernetes Engine (GKE) in 30 minutes, leveraging Strea…
Anthropic paused a planned billing change for its Claude Agent SDK on June 15, which would have separated programmatic usage from subscription pools and charged API rates. The change would have ended an implicit subsidy …
SpaceX agreed to acquire AI coding startup Cursor for $60 billion in stock on June 16, 2026, raising concerns about using the same AI stack for both code generation and review. Industry data shows a 14x increase in AI-ge…
MagmaLabs engineer Ben Sheldon describes how he built an automated loop that fixed a failing spec in Forge, the company's Rails 8 virtual employee platform, without manual prompting. The concept, called 'loop engineering…
The US Commerce Department ordered Anthropic to disable its Fable 5 and Mythos 5 AI models for all foreign nationals, citing national security concerns over a jailbreak that could weaponize the models' vulnerability-hunt…
A Fortran programmer documented using LLMs and agentic workflows to write GPU-accelerated Fortran code, with posts on software architecture, OpenMP target offloading, and standard parallelism. The programmer plans to sha…