Large language model (LLM) news — GPT-4, Claude, Gemini, Llama, Mistral and the latest research on training, fine-tuning, RLHF, and deployment of LLMs.
Researchers introduced LoHoSearch, a new benchmark of 544 human-verified questions across 11 domains designed to test long-horizon search agents beyond the human difficulty ceiling. The benchmark, built from a knowledge …
Researchers fine-tuned three small language models—Phi-3-mini, Qwen2.5-3B, and Mistral-7B—using QLoRA on biomedical claim verification datasets, finding that Mistral-7B outperformed GPT-4o and GPT-5 by up to 12% F1 at a …
Researchers have demonstrated that Direct Preference Optimization (DPO) simplifies the fine-tuning pipeline for large language models while improving computational efficiency and achieving competitive performance. The em…
Researchers at arXiv have proposed SafeLLM, an extraction-based method for retrieval-augmented generation that eliminates hallucinations by directly selecting relevant sentences from source documents rather than rewritin…
Linear launched coding sessions that allow users to run agentic coding workflows directly within the platform, enabling automated pull request creation and code review through Claude Code or Codex. The feature, available…
Apple debuted the third generation of Apple Foundation Models (AFM) on June 8, 2026, unveiling a family of five models spanning on-device and server deployments, including the 20-billion-parameter, natively multimodal AF…
Researchers released a multilingual vaccine benchmark called VaxEval containing 1,886 multiple-choice questions covering 14 vaccines in English, Spanish, and Chinese, drawing from sources including the World Health Organ…
Sophia NLU v1.2, an on-device, low-compute natural language engine built in Rust, now runs voice assistants without internet or a GPU, using only 160MB of RAM. The upgrade adds entity clarification, state persistence, an…
MandoCode, a new open-source AI coding agent built on .NET and Ollama, launched today as a local-first alternative to cloud-based assistants. The tool, which requires no API keys and runs entirely in the terminal, can re…
The MADSys Lab at Tsinghua University’s KTransformers project enables frontier-class MoE models like DeepSeek-R1 671B to run on commodity hardware with a CPU-GPU hybrid inference stack, achieving 286 tokens/s prefill on …
A solo developer shipping 7 Mac apps uses VS Code with rust-analyzer, a three-tier AI stack of local Ollama autocomplete, Gemini for quick lookups, and Claude for complex Rust bugs, plus iTerm2 and a strict "commit befor…
Indian IT shares fell for a seventh consecutive session, with the Nifty IT index closing at its lowest level since May 15 after dropping 1.6%, as investor concern over Anthropic's new Claude Fable 5 AI model intensified.…
Google released A2UI (Agent-to-User Interface) in late 2025, an open protocol that enables LLMs to compose user interfaces declaratively rather than selecting pre-built components. The protocol transmits UI as a stream o…
LangChain launched LangSmith Sandboxes, a new infrastructure product that gives AI agents their own isolated computers with filesystems, shells, and package managers. The product addresses security risks from agents runn…
By early 2026, "vibe coding" — describing software in natural language and accepting AI-generated output — had become a mainstream development workflow, with Collins Dictionary naming it Word of the Year and tools like L…
Apple announced five third-generation Apple Foundation Models during WWDC26, including on-device models AFM 3 Core and AFM 3 Core Advanced, and server-based models AFM Cloud, ADM 3 Cloud (Image), and AFM 3 Cloud Pro. For…
A developer has completed PuskesmasAI, an offline-first Progressive Web App that uses AI-powered medical triage to assist community health workers in rural Indonesia. The app runs entirely on-device with no internet conn…
LearnThat, a new remote HTTP MCP server, enables coding agents to integrate small learning exercises into development workflows by adding coaching tools such as level assessments, micro-challenges, and short-answer gradi…
RAG (Retrieval-Augmented Generation) enables AI applications to retrieve relevant information from external data sources and use that context to generate accurate responses, overcoming the limitation of large language mo…
A developer replaced brittle regex and CSS selectors with a local LLM to extract product data from messy legacy HTML, achieving reliable results at a fraction of the cost of cloud-based models. The approach uses Ollama's…