LLM-Typesafe 0.1a0 is a new plugin for LLM that adds support for TypeSafe AI's Jev model, enabling yes/no and choice questions with scoring.

LLM-Anthropic released version 0.29, which adds support for Claude Opus 5.5. This update provides users with the ability to interact with the Claude Opus 5.5, a large language model.

LLM 0.36 has been released with two new OpenAI models, gpt-6-sol and gpt-6-luna, and several improvements. Model plugins can now declare support for single-turn prompts, and reasoning traces in logs are now formatted with <details><summary> tags.

OpenAI and Anthropic have released new models, GPT-6 Sol, GPT-6 Luna, and Claude Opus 5.5, with significant price reductions. GPT-6 Luna is one of the cheapest models OpenAI has released, and Opus 5.5 addresses user feedback on its communication style and is cheaper per token than Opus 5.0.

Relay is a self-hosted LLM gateway with smart routing and request pacing. It classifies incoming requests, decides on model capabilities, and routes them across configured providers while accounting for limits and available capacity. The gateway is lightweight, ships with routing classification models, and has a built-in classifier that runs in the single digit millisecond range.

ai·rete·rag is a tool that combines a Rete rule engine with a language model to provide auditable decisions. It runs the rule engine first, then uses the LLM to generate a plain-English explanation of the decision. The tool has a visual editor and supports forward chaining and nested rules.

GPT-6 Astra has been used to reduce research time and cost by half for Parallel's agents, demonstrating improved efficiency in labor-market data synthesis.

OpenAI released GPT-6 Sol and Luna, two AI models that bring frontier intelligence to everyday work with different capability and cost balances.

OpenAI has improved GPT-6's prompt caching with higher cache hit rates, new diagnostics, breakpoints, and controls to reduce latency and costs for AI agent builders.

OneBid is a unified auto-bidding foundation model for diverse oCPX advertising scenarios, built on top of the Decision Transformer (DT) model. It extends DT with two atomic signals and a sequence-level Mixture-of-Experts architecture to learn from heterogeneous logs and adapt to scenario-specific deployments. OneBid is validated through online A/B tests and delivers improved performance in real-world scenarios.

A new plugin, llm-keys-ui, has been released. It allows users to securely manage API keys for LLM projects without pasting them directly into the ChatGPT app. The plugin uses Codex to run commands and retrieve key URLs for local network or Tailscale device IPs.

datasette-auth-github plugin version 1.0 released, includes fix for session cookie expiration issue, tested on Datasette 0.65.x and 1.0ax, and is now considered stable.

Chronicle is a new tool for regression testing of LLM agents. It records agent runs at non-deterministic boundaries as immutable envelopes and replays them from the record. This allows for the creation of regression tests that can catch faulty code and pass on guarded and benign changes. Chronicle is available on GitHub.

Researchers studied the impact of planning and release control on the performance of stateful LLM agents in retail and airline experiments. They found that prewritten plans improve success by 7.17 percentage points and reduce erroneous acceptance by 61%, with the planning component dominating at low liability and the verifier's benefits dominating at high liability.

Claude Code now supports using AGENTS.md files instead of CLAUDE.md if the latter is missing. This change is built on Claude Code mods, a way to customize the harness. This update enables users to build custom project instructions, with the source code available for modding.

TurnPanel is a local-first AI workspace that allows agents to operate within a computer, orchestrate other tools, and preserve context. It supports multiple models, including Codex and OpenRouter, and has a mini-app SDK for building custom tools.

Visual Perception Engine (VPEngine) is a modular framework for efficient GPU usage in robotic vision tasks, enabling parallel execution of multiple machine learning models. It uses a shared foundation model backbone and task-specific heads, achieving up to 3x speedup and efficient GPU utilization.

RLLBC-Lib is an open-source code library for reinforcement learning and learning-based control, aiming to simplify the learning process for students and developers. It provides a comprehensive library of tabular RL approaches and a deep RL library, as well as implementations illustrating core RL principles.

WARD is a runtime-adaptive Vision Transformer framework for dependable edge AI that adapts to changing power budgets, reliability requirements, and input distributions. It uses channel-wise subnetwork partitioning, reliability-aware continual learning, and dynamic operating-mode scheduling to optimize performance, fault tolerance, and adaptation.

A new open-source machine-learning library, TuiML, is designed specifically for AI agents. It provides native algorithms and machine-readable metadata for self-discovery, composition, and validation of workflows. This library is competitive with Weka and scikit-learn, and is intended to allow agents to read, extend, and operate machine learning autonomously.