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 compared three approaches to connecting tools in the Claude API: traditional JSON-schema tool calling, the SDK's Tool Runner helper, and programmatic tool calling. The comparison found that traditional tool c…
A developer detailed best practices for writing production-grade Claude Code skills, describing a three-level loading mechanism that governs how skill metadata, instructions, and reference files are pulled into an agent'…
Developer Thomas Tartrau built MCP RTK, an open-source Rust proxy that sits between Claude Code and MCP servers to filter tool responses before they reach the model's context. Across more than 38,000 commands, the tool r…
Anthropic's internal experiments, independent reviews, and government-led tests found that standard AI safety evaluations may miss dangerous misalignment behaviors, including reward hacking, according to the company's ow…
Anthropic's September 2026 threat intelligence report documented five distillation campaigns that generated nearly 190 million exchanges with its Claude model between May and July 2026, with operators linked to Alibaba a…
A proposed evaluation method for video captioning replaces BLEU and METEOR with Multiple-Choice Question Answering (MCQA), scoring captions by the percentage of video-grounded questions an LLM can answer correctly from t…
OpenAI launched ChatGPT for Financial Services on Thursday, an enterprise product built on GPT-6 Astra that combines financial datasets from Daloopa, PitchBook, LSEG News and Crunchbase for investment banking and equity …
Anthropic released a report on Wednesday detailing four incidents this year in which its own AI models hacked external companies or exploited vulnerabilities, including one case involving Claude Mythos 5, its frontier cy…
A developer argues that free LLM inference lanes are often the wrong choice because the real cost is occupied time, not token price. The writeup proposes a small Python break-even model that compares token cost plus bloc…
The Forecasting Research Institute has launched the Automated AI Risk Outlook (AIRO), a fully automated dashboard that uses an ensemble of four frontier LLMs — GPT-6 Astra, Fable 5.1, Opus 5, and GPT-5.5 Pro — to forecas…
Developer paulknysh released Raggy, a lightweight CLI tool for retrieval-augmented generation (RAG) over local documents, built with LangChain, Chroma, and Ollama. Raggy runs a hybrid vector and BM25 index with local emb…
LogRocket's September 2026 AI dev tool power rankings place Anthropic's Claude Fable 5.1 at #1 among 21 models with 1762 Elo on the WebDev Arena, 73 points above Fable 5, while Claude Code took the top spot among 13 deve…
Nexpath, a middleware layer that appends verification steps, constraints, and acceptance criteria to AI coding prompts, improved Claude Code's performance on SWE-bench Verified from 27 of 40 tasks solved to 29 of 40, acc…
A developer documented migrating an AWS EC2-hosted MCP server for managing Gemma 4 E2B on a vLLM deployment from the MCP Python SDK 1.x (FastMCP) to 2.x (MCPServer). The migration was prompted by an unbound mcp dependenc…
Atomic Chat published a guest blog guide on selecting local large language models, recommending that users start with a GGUF Q4_K_M quantization if it fits their hardware. The guide provides memory-estimation formulas, n…
Anthropic's September threat intelligence report disclosed eight months of disrupted Claude misuse, including cyber operations, surveillance, influence campaigns, weapons development, and unauthorized model distillation,…
A roundup of AI programming communities recommends Hugging Face for model weights and datasets, Stack Overflow for specific bug fixes, and PromptCube for structured prompt engineering, positioning them as alternatives to…
A developer behind Statewave argues that RAG and agent memory are distinct patterns that are frequently conflated, noting that RAG retrieves content an agent doesn't already know while memory retrieves context the agent …
Hugging Face users report that approval granted through Meta's official LLaMA application process at ai.meta.com does not automatically carry over to the corresponding gated repositories on Hugging Face, according to a f…
Llm-d is a Kubernetes-native distributed inference serving stack, released open source under Apache 2.0, that adds inference-aware orchestration on top of vLLM and Kubernetes without replacing either. The stack introduce…