EverMind-AI/Raven

One Surface, All Agents: Raven generates DAGs and orchestrates multiple specialized agents for complex tasks.

Raven is the harness of harnesses, built for recursive self-improvement (RSI). As a Host Agent, it brings built-in and third-party agents together to carry out complex tasks. Its modular architecture supports iterative improvement of Raven's own harness: proposing changes to how agents plan and act, evaluating those changes, and adopting improvements that pass validation. Powered by EverOS, Raven carries memory and context across sessions to support this process.

Built-in Agents: Raven-Research, Raven-Code, Raven-Design, and Raven-Oncall support research, coding, visual design, and unattended workflow automation.

Raven is pre-alpha. Interfaces and configuration may change quickly.

Raven's Performance on the Multi-Agent Orchestration Benchmark

These are three complete projects delivered by Raven. In each case, Raven drove a team of specialized agents from the initial brief or objective through execution to a complete set of final deliverables.

The brief came from a person; Raven completed the entire project. Working autonomously for about 4 days, Raven completed 42 rounds of planning, development, and verification to build a playable first-person shooter in Godot 4, centered on an arena boss fight. The full deliverable includes the game, its poster, presentation, and website, all produced by Raven.

AI that improves AI, with the entire project completed by Raven. Given a task and evaluation criteria it cannot modify, Raven RSI independently plans each round, writes code, runs experiments, and evaluates the results. In nanochat pre-training experiments, it completed 172 training runs across 7 rounds without a single crash, reducing val_bpb by 5.8% within the same 20-minute, single-GPU budget. The same process reduced overshoot in a dam-break simulation by three orders of magnitude and completed an FEA limit-load search in 8 rounds of bisection. The complete deliverable includes the experimental results, visualizations, poster, presentation, and project website, all produced by Raven.

One raven, a whole flock of specialists. Raven completed the entire project. Its launch kit brings together a browser-based physics mini-game, a 16-slide product overview, posters in English and Chinese, and the README you are reading now, all produced by Raven.

More showcases

Raven's modular architecture is designed for harness self-evolution and subagent creation. Its four built-in agents deliver state-of-the-art (SOTA) performance in their respective domains, combining reusable harness components with domain-specific tools, skills, and agent loops. Raven can delegate a focused task to a single agent or orchestrate multiple agents within a shared workflow. The harness they share is refined by the Raven Evolver, a separate tool that consumes Raven as a library and evaluates candidate harness changes against benchmarks; it develops the agents rather than running inside them.

All four agents are built in and ready for orchestration out of the box.

Raven-Research enables autonomous deep research for complex questions, literature reviews, and technical analysis. It delivers clear, structured reports with traceable sources, helping users understand unfamiliar domains, compare alternatives, and make informed decisions.

Raven-Research's performance on the DeepResearch Mixed benchmark

Raven-Code enables agentic software development, turning requirements into working, tested code. It supports feature implementation, debugging, refactoring, data processing, and data analysis, helping users build new capabilities, resolve issues, and improve code quality while following their project's conventions.

Raven-Code's performance on coding benchmarks

Raven-Code tops on DataAgentBench for data analysis (2026-08-24 Live)

Raven-Design performs visual design, turning ideas and content into polished visual deliverables. It creates PowerPoint slide decks, brand assets, charts, diagrams, and web interfaces, refining layout, typography, and visual consistency to help users communicate clearly and bring their ideas to life.

Raven-Design tops on PresentBench for slide generation

Raven-Design's performance on visual design benchmarks

Raven-Oncall enables unattended workflow automation for experimentation, optimization, and continuous monitoring. It autonomously manages workflows from start to completion, sustaining progress over hours or overnight, delivering results, and involving users only when human judgment is needed.

Raven-Oncall significantly outperforms Claude Code on both quality and cost for AI4AI tasks

Raven-Oncall significantly outperforms Claude Code on both success rate and cost for AI4S tasks

Raven is built for this from the ground up. Its agent loop is split into four decoupled strategy modules — Memory for what a turn gets to see, Planning for how it approaches the work, Capability for which tools an iteration exposes, Action for what to do next and for judging it before it runs. A Curator keeps rewriting those four seats: a setting, or a small piece of judgement code written for that agent. What it changes belongs to that agent alone, and once installed the agent runs on it.

A Curator changes more than the prompt — the tools and outside services it reaches for, the skills and procedures it follows, and its own judgement at each point can all be replaced. It keeps going round after round — you put it to work, you say what was wrong, it reworks — until you are satisfied, until it has nothing left worth changing, or until the round budget runs out. Two things hold the quality: nothing is installed before it is verified, and a failed check sends it back; and a round's signals normally reach it with the reference answer stripped out, which limits how directly the answer is exposed. The Curator is experimental: it ships with the repository rather than the installed package.

A Persona is the first thing it builds. Describe the assistant you want and the Curator assembles one: a lead role that talks to you, and specialists drawn from the agents you already have. What you asked for lands in that assistant's harness — its division of work, the tools it may reach for, and the checks it must pass before it acts.

Describe what you need once. Raven assembles the assistant, keeps improving it while you work, and afterwards a sentence is enough to put it to work again.

Raven can connect to and orchestrate agents via ACP, CLI, or OpenAI-compatible APIs, with presets for 13 third-party agents to simplify setup, task delegation, and coordination across shared workflows. Try these agents in Raven through a unified interface!

• 📣 2026-09-30 — Raven's technical report is #1 Paper of the day on Hugging Face.

Let your own agent install Raven for you. Copy this prompt into any agent that can read a web page and run shell commands, such as Claude Code or Codex:

Read https://evermind-ai.github.io/Raven/quick-start/ and follow it to install Raven, or to update it if it is already installed.

Linux, macOS, or WSL2:

curl -fsSL https://raven.evermind.ai/install.sh | bash

Native Windows PowerShell:

irm https://raven.evermind.ai/install.ps1 | iex

Windows PowerShell 5.1, the version built into Windows, cannot follow that URL's redirect and stops with (308) Permanent Redirect. If you see that error, use the direct installer URL instead:

irm https://raw.githubusercontent.com/EverMind-AI/Raven/refs/heads/main/install.ps1 | iex

Or run it in a container, with nothing on the host but Git and Docker:

git clone https://github.com/EverMind-AI/Raven.git
cd Raven
docker compose -f docker/docker-compose.yml up

Then open http://localhost:18793 in your browser. Prerequisites are Git and Docker with Docker Compose. See docker/README.md for what comes up, how the page signs itself in, and how to run it behind a remotely exposed port.

Or install from a source checkout, to develop against the code or to run what has not been released yet:

git clone https://github.com/EverMind-AI/Raven.git
cd Raven
./install.sh

Run as a file, install.sh installs that checkout in editable mode: raven and its bundled plugins link back to your tree, and the TUI bundle and the served page are built from it. A piped run installs the published wheel even from inside a clone, so that a one-line install never picks up whatever a working tree happens to contain. Set RAVEN_LOCAL_SRC=<dir> to force the editable install through a pipe.

The agent products ship with raven itself: a wheel carries the agents/ product tree and copies it out to your raven home on first use, and a source checkout reads the tree in place. Setup asks, for each product, whether it runs on the model it is tuned for, which needs a key of its own, or on this raven's LLM. Nothing is registered: a folder is on the roster because it is there. See agents/README.md.

Learn more about Raven on the documentation site.

Read the documentation

Raven's WebUI brings conversations, multi-agent collaboration, and workspace management into your browser. Chat with agents, follow task progress, inspect files and outputs, and browse memory and skills in one place.

raven web

The command opens the WebUI in your browser and keeps Raven running in the background. Use raven web --stop to stop the background service.

New task: one composer, with skills, playbooks, knowledge and memory a click away.

Subagents: every connected agent in one roster, built-in and third-party alike.

EverMind connects memory research, production-ready products, and practical integrations into one open-source ecosystem.

Together, these projects form EverMind's research-to-runtime stack: methods and benchmarks become reusable memory infrastructure, products, and agent integrations.

Issues and pull requests are welcome. Start with the developer workflow, follow AGENTS.md for repository rules, and use GitHub Discussions for design conversations.

Apache License 2.0

If you use Raven in your research, please cite the technical report:

@misc{evermind2026raven,
  title         = {{Raven: The Harness of Harnesses for Composable Agentic Intelligence}},
  author        = {{EverMind AI}},
  year          = {2026},
  eprint        = {2609.33439},
  archivePrefix = {arXiv},
  primaryClass  = {cs.AI},
  url           = {https://arxiv.org/abs/2609.33439}
}