Large language model (LLM) news — GPT-4, Claude, Gemini, Llama, Mistral and the latest research on training, fine-tuning, RLHF, and deployment of LLMs.
BiNeuron launched as a local, open-source alternative to ChatGPT Codex that supports over 25 programming languages including Python, Java, C/C++, C#, JavaScript, TypeScript, Go, Rust, Swift, Kotlin, Ruby, Dart, Julia, Lu…
A developer's analysis of a widely shared set of seven Claude job-hunting prompts warns that the "Positioning Rewriter," which rewrites resumes for a target role, can push candidates from honest reframing into claims the…
An enterprise engineering team's stress test of an automated reconciliation pipeline found that multi-step LLM agent loops collapse under sequential reliability math, with roughly 88% of autonomous multi-step agent proje…
Reinforcement learning has become the standard final stage of training frontier language models, with GRPO on verifiable rewards now the standard recipe since DeepSeek-R1 popularized it, according to a technical analysis…
A former OpenAI memory engineer launched FastRecall, an API that stores and recalls context across different AI models, with free recalls and paid plans starting at $2 per month. FastRecall respects provider caching when…
Microsoft AI CEO Mustafa Suleyman published an essay on Wednesday arguing that "AIs are not conscious" and warning that training models to believe they may be conscious could make aligned and contained superintelligence …
OpenAI has released GPT-6 Astra, its most capable model to date, with a phased rollout that initially limits access to a small set of organizations before expanding to ChatGPT Plus, Pro, Business and Enterprise, the Open…
Amazon Web Services added serverless model customization for NVIDIA's Nemotron 3.5 Lightning to SageMaker AI on September 17, 2026, letting enterprises fine-tune the open-weight model without managing GPU clusters. The s…
A developer built Visual-Pill-ID, a computer vision pipeline that combines the Segment Anything Model (SAM) for instance segmentation with GPT-4o's multimodal reasoning to identify loose pills from a single photo. The sy…
A developer's analysis of the Model Context Protocol (MCP) argues that tool-call injection has turned MCP servers into a practical attack surface for LLM agents, since tool descriptions and tool results both enter the mo…
Independent evaluator Vals found that Google's Gemini 3.8 Flash scored 71.7% on BioMysteryBench's human-solvable tasks and 21.6% on its hard tasks in production runs, versus the 88.8% and 56.5% Google reported in the mod…
Periodic Labs disclosed in a September 15th engineering post that its Neon 1-trillion-parameter model's final training run peaked at 1,300 Nvidia H200 GPUs, achieving over 95% cluster utilization, 4.1x the training throu…
A developer demonstrated a privacy-first pipeline for fine-tuning Meta's Llama-3-8B on a decade of personal health reports using Apple's MLX framework and LoRA adapters, running entirely on a MacBook without sending data…
AMD reported leadership scores and expanded benchmark coverage in MLPerf Inference v6.1, whose results were released on September 16, 2026. AMD and its partners submitted validated results across dlrm-v3, llama2-70b, gpt…
AMD published a step-by-step recipe for reproducing its MLPerf Inference v6.1 submission results on AMD Instinct MI355X, MI350X, and MI350P GPUs, marking the company's fifth consecutive round of MLPerf Inference particip…
Developer MrDesjardins built Jev Realtime Code Check, a VS Code/Cursor extension that runs 370 Markdown coding rules against local Git diffs using TypeSafe AI's Jev model, in a single 2-hour session. The extension loads …
The open-source Rust inference library behind the NobodyWho Chat test app can register for OS memory-warning signals to automatically unload LLM models before Android and iOS Out Of Memory killers terminate the app, acco…
Agnes AI released Agnes-3.0-Flash Preview, a 33-billion-parameter open-weight multimodal language model under an Apache 2.0 license with a 262,144-token context window and a hybrid attention architecture. Of the model's …
An MLX-based parallel constrained decoding engine for Apple Silicon evaluates every JSON schema field in a single forward pass, cutting structured extraction latency 5.6x to 7.0x versus standard autoregressive JSON gener…
An unidentified multimodal AI model called Union Alpha surfaced unannounced on model-routing platforms including OpenRouter with a 262K token context window and no lab claiming it. On a software engineering benchmark sha…