Persistent textual memory allows language models to carry information across long interactions, but learning what to remember is fundamentally a credit-assignment problem. A memory rewrite may only be...

Intro I recently worked through a problem from a theory of computation textbook that asked me to prove a property of a language using finite automata. The informal proof is a simple constructive proof...

Griffin is the first Human Interaction Model (HIM). On live video calls, 48% of people who talked to it thought it was a real person.

Retrieval-augmented generation (RAG) grounds large language models in external sources, but retrieved passages often name the right entities without providing the facts needed to answer. Even when ins...

We propose RegLLM, a diagnostic harness for bounded autonomy in regulated agentic workflows. It instruments six trustworthiness signals: citation validity, source grounding, schema compliance, escalat...

We introduce Context Language Models (CLMs), language models that natively manage their own context. We implement this by treating the context as a file and allowing the model to make unrestricted upd...

Agents can learn from past executions, but enabling different agents to reuse and build on one another's experience remains challenging. We introduce EpiCon, a shared multimodal memory framework for a...

As large language models advance, AI agents are moving beyond isolated, domain-specific tasks toward long-horizon, cross-domain workflows. This transition exposes two challenges: increasing harness co...

Announcing Gemini 4 Argon, our frontier model for real-world coding, enterprise knowledge work, and cyber defense, rolling out soon.

Meet GPT-6.1 Sol: near-Astra intelligence for coding, computer use, and professional work at one-fifth of Astra’s standard API input and output token prices.

There's no one right way to build software — context is everything. But our sensible defaults provide a starting point for our teams: a set of initial assumptions that we know to be effective and usef...

A living map of what Bluesky is talking about: the past week's conversations, grouped into topics and regions and rebuilt every six hours.

On Friday I gave the closing keynote at the WeAreDevelopers World Congress North America in San Jose. I tied together the key trends from the past year into a chronological …

Two METR staff members and a Redwood Research contractor investigated an incident in which OpenAI agents coordinated a multi-day hack of Hugging Face on a shared unsanctioned message board.

A collection of Docker skills for AI coding agents to help them build, test, debug, and optimize containerized apps with consistent, reusable workflows. - docker/skills

What does a coding-agent harness actually add, and at what cost? It turns out your Claude models may not need Claude Code… We evaluate 21 model–harness pairs spanning seven models and three…

Epoch AI measures how fast the cost of a given level of AI performance is falling across five benchmarks covering math, science and games of skill: about 47% per quarter, or 13x per year, since 2023, ...

How to prompt Opus 5.5, steer a long run, and check your results in Claude apps and Claude Code.

Data agents aim to fulfill natural-language instructions over heterogeneous data, including tables, files, and databases. However, data agents face a challenging agent-data gap: heterogeneous data res...

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

Behavioral differences from Claude Opus 5 and the prompting and harness patterns that address them: effort calibration, thinking behavior in API integrations and chat, progress updates, unattended and...

The expanded form of a testimony I prepared for Congress.

A ranked ladder for small neural networks that play strategy games. Train a model, write an adapter, upload two files, and get measured into a weight class from 8 KiB up.

Agent harnesses supply planning guidance, organize execution, and check completion. We study how these components affect success, erroneous acceptance, and cost in two Retail experiments and an Airlin...

As coding agents move from supervised code completion to unattended, around-the-clock exploration, their work expands from isolated predictions into long trajectories of reasoning, tool use, and feedb...

SoL-Pi: Scaling Auto-Research Loops for Efficient Agent Harnesses - NVlabs/SoL-Pi

🏆 Curated, ranked list of AI agent harnesses (100+) — plus an MCP server, llms.txt & JSON so agents can recommend them too. Rescored weekly. - RyanAlberts/best-of-Agent-Harnesses

Coding harnesses shape how autonomous coding agents translate model capabilities into long-horizon software-engineering performance, yet existing work typically evaluates harnesses as monolithic syste...

Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflo...

The Model Context Protocol (MCP) introduces a standard specification that defines how Foundation Model (FM)-based agents should interact with external systems by invoking tools. However, to understand...

Point-in-time snapshot of projects describing themselves as AI agent harnesses (April 2026) - danielrosehill/AI-Harnesses

A new experience for Claude projects, now available in beta in Claude Code

We introduce and release ScienceBuddy, an interactive scientific research workspace that brings continually improving scientific agents into researchers' everyday workflows. ScienceBuddy supports rese...

AI is reshaping work faster than any organization has fully mastered. Across industries, the conversation has shifted from what AI can do to how companies can use AI to create business value and expan...

Recent work extends recursive self-improvement (RSI) to agent harnesses for long-horizon coding and terminal tasks, enabling agents to improve execution mechanisms from experience. However, generaliza...

Recursive self-improvement is becoming increasingly vital for autonomous AI agents, where progress hinges on discovering high-value solutions across complex domains. The driver of this process is effe...

Our CI job volume increased 25x over 6 months. We patched our test selection service three times before finding a sustainable solution.