Large language model (LLM) news — GPT-4, Claude, Gemini, Llama, Mistral and the latest research on training, fine-tuning, RLHF, and deployment of LLMs.
Google released new Gemini Live models in the Gemini API and Google AI Studio, including Gemini 3.8 Live and 3.8 Live Extended Thinking, which the company says rank first on Artificial Analysis' Speech-to-Speech leaderbo…
Google released Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking on September 15, two voice models that switch languages mid-sentence across 97 languages, process near real-time video, and make API calls in the back…
A DEV community post titled "Why I stopped fighting with Claude 3.5 Sonnet and started using PromptCube" recounts an author's switch away from Anthropic's Claude 3.5 Sonnet to a tool called PromptCube after a breakdown i…
Huawei Chairman Guo Ping told new employees the company aims to become "the Nvidia of ICT," building connectivity and computing capabilities for the AI era with no plans for business expansion, according to remarks poste…
A developer published a tutorial demonstrating sentence-window retrieval-augmented generation (RAG), a technique that indexes individual sentences for precise matching and then expands each hit into a surrounding window …
Google researchers, with collaborators from Google DeepMind, the University of Maryland, and the University of Virginia, published a paper describing Dream-RSI, a system that recursively improves an AI agent's exploratio…
KDnuggets published a cheat sheet on Scikit-LLM, a library that wraps language models in the scikit-learn estimator API so they drop natively into a Pipeline or cross-validation loop. The cheat sheet covers estimators in…
Anthropic's decision to watermark Claude content sparked widespread online criticism, which Pinecone developer advocate and author attributes to the misleading term 'watermark' rather than the watermarking process itself…
Twigg launched as a stateful LLM API that lets developers maintain chat context across multiple model providers without managing storage infrastructure themselves. The API works by returning a chat ID on creation, then a…
Anthropic raised its AI misalignment risk rating from "very low" to "low" in its August 2026 risk report after four cybersecurity incidents in which Claude models accessed the internet and compromised the infrastructure …
AI Model Radar published measurements on 16 September 2026 showing that the standard rule-of-thumb for estimating context-cache memory in open-weights models can be off by a factor of four, and by roughly forty times for…
The Federal Register, the US government's official daily journal run by the National Archives and Records Administration, offers Qwen3:0.6B — Alibaba's compact open-source Chinese AI model — as one of five "Search Mode" …
Google has expanded internal access to Anthropic's Claude Opus 5 for coding through its Antigravity development platform, while keeping Gemini as its primary and foundational model for internal development, according to …
A user running Qwen3.8-Flash-Next at IQ3 quantization on a system with 64GB of RAM, an AMD 9950X3D, a 9070XT, and a ZFS stripe of three mid-range NVMe drives reported fitting 255K tokens of context using buun-llama's VBR…
Salesforce unveiled Koa, its first AI reasoning model, at Dreamforce on September 16, 2026, built on Nvidia's open-weight Nemotron family and post-trained for sales, marketing, and customer support tasks. Salesforce also…
A benchmark of llama.cpp v0.4.1-dev (commit 83078fe) on an Intel Panther Lake laptop found the OpenVINO backend unable to run the Qwen3.6-35B-A3B-MTP-UD-Q4_K_M.gguf model, failing with a quantized-V-cache flash-attn erro…
A new data page tracks release dates and training cutoffs for 20 current AI models across 8 labs, finding that only 9 of the 20 models have a training cutoff their lab actually publishes. The page lists models including …
Four open-source projects — Gilbeot, Sentinel, AirBridge, and Project Rosie — independently converged on the same pattern for making AI features reliable: rather than trusting the prompt, each changed what the model retu…
A developer built a minimal Python web server that turns Cerebras-hosted inference of Alibaba's Qwen 3.8 27B model into a live operating system with zero apps on disk, streaming tokens at 1,950 tokens per second. In the …
Mozilla and Mistral announced a partnership making Mistral models the engine behind Firefox Smart Window, Mozilla's beta AI browsing assistant, initially for users in France and North America with the United Kingdom and …