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 announced at its Google Cloud Security Summit that its autonomous code-security agent CodeMender, in preview since July, is now used by hundreds of customers to find and fix vulnerabilities at machine speed, addin…
Vercel's AI Gateway now serves Moonshot AI's Kimi K3.1 Pro and Kimi K3.1 Pro Highspeed through additional Zero Data Retention (ZDR) providers, where Moonshot AI had been the only provider for both models. Both models off…
A development team engineered an AI agent architecture that adds persistent long-term memory to otherwise stateless LLM sessions, mapping code configurations and runtime variables into a localized semantic layer. The des…
Moonshot AI, led by co-founder and chief executive Yang Zhilin, published the weights for its 2.8-trillion-parameter Kimi K3.1 Pro model on Hugging Face on November 4th, 2026, under a Kimi-specific license, fulfilling a …
A reference implementation built on the Strands Agents SDK and Amazon Bedrock AgentCore cut the tokens each script adds to context by about 95 percent through a single write-and-execute tool, and shrank the system prompt…
Indegene Limited (BSE: 544172, NSE: INDGN) built a production-grade, LLM-powered Data Quality Agent inside its NEXT Unified Data Platform on AWS that generates context-aware data quality rules, validates them, and produc…
A developer at Lunil, a marketplace for buying and selling digital assets such as SaaS products, websites and domains, built a synchronous listing-moderation pipeline using Next.js App Router route handlers, the DeepSeek…
Tencent Holdings Ltd. released Hy4, a new frontier large language model that adds computer-use capabilities for the first time, combining visual understanding, reasoning and tool-calling for long-horizon agentic performa…
Confluent released its Q4'26 Data Streaming Platform update, adding Query as a fifth DSP pillar alongside Stream, Connect, Process, and Govern, with Lightning Tables at its center to make Kafka streams queryable in milli…
A developer reports that a deep-research project connecting his mother's health-preserving pedagogy research to machine learning methods has already produced its core findings, seven weeks before the brief was formally r…
MIT researchers demonstrated that OpenAI's GPT-6 Astra can read and deliberately rewrite the weights of other AI models without traditional training, boosting a Qwen model's math accuracy from 7.9 percent to 48.1 percent…
A TIL post outlines six techniques for reducing large language model inference costs: prompt caching, model routing, batch APIs, semantic caching, context trimming and smaller or distilled models. The post cites Anthropi…
Datasette-llm 0.1a10 was released on 4th November 2026, fixing a bug reported in pull request 11 on the datasette-llm GitHub repository that affected the agent.datasette.io demo site. The bug caused the set of available …
Anthropic released FlashProve, an open-source research prototype that uses zkSNARK cryptography to prove which model generated an AI response without revealing its weights, generating proofs at roughly 110 tokens per sec…
A developer at wrightfunctions.com outlined four techniques for cutting LLM inference costs by routing requests between cheap and expensive models rather than sending every query to the priciest one. The writeup cites An…
A paper submitted to arXiv on 3 November 2026 reports that frontier open-weight language models lose little to no task accuracy across a broad range of long-horizon and long-input tasks when contexts are kept short throu…
MariaDB on Wednesday added four agent-focused capabilities to Enterprise Platform 2027: AI Retrieval, AI Plugins, Agent Memory and Semantic Cache. AI Retrieval converts natural-language questions into read-only SQL queri…
A 4-billion-parameter looped (latent-recurrent) language model called Dory1.0-4B was introduced, built on a normalized Transformer (nGPT) backbone with interleaved sliding-window and full-attention layers. Dory1.0-4B off…
Tencent announced Tencent Hy4 on November 4, 2026, a 770 billion parameter Mixture-of-Experts large language model that activates 49 billion parameters per token and ships under the Apache License 2.0, with downloadable …
MIT CSAIL researchers Rikiya Takehi, Jayden Teoh and Alex Zhang reported in the paper "Sparks of Weight-Space Programming" that a frontier AI model, GPT-6 Astra, directly edited 131,072 parameters — 0.016% of the 800 mil…