AutoGPT ships as two binaries in a single crate: autogpt (the agent runner) and orchgpt (the orchestrator). Choose your preferred installation method below.
The fastest way to install both binaries:
cargo install autogpt --all-featuresAfter installation, verify both binaries are available:
autogpt --version
orchgpt --versionDownload the latest pre-built executables directly from GitHub Releases without a Rust toolchain:
| Platform | Download |
|---|---|
| Linux x86-64 | autogpt |
| Windows x86-64 | autogpt.exe |
Run AutoGPT inside an isolated container, no Rust installation required.
AutoGPT agent container:
docker run -it \
-e GEMINI_API_KEY=<your_gemini_api_key> \
-e PINECONE_API_KEY=<your_pinecone_api_key> \
-e PINECONE_INDEX_URL=<your_pinecone_index_url> \
--rm --name autogpt kevinrsdev/autogptOrchGPT orchestrator container:
docker run -it \
-e GEMINI_API_KEY=<your_gemini_api_key> \
-e PINECONE_API_KEY=<your_pinecone_api_key> \
-e PINECONE_INDEX_URL=<your_pinecone_index_url> \
--rm --name orchgpt kevinrsdev/orchgptBuilding from source gives you access to all development features and the latest unreleased code.
1. Clone the repository:
git clone https://github.com/wiseaidotdev/autogpt.git
cd autogpt/autogpt2. Build and run the agent CLI:
cargo run --all-features --bin autogpt3. Build and run the orchestrator:
cargo run --all-features --bin orchgpt4. Or build an optimized release binary:
cargo build --release --all-features
# Binaries land in: ./target/release/autogpt and ./target/release/orchgptTo embed AutoGPT in your own Rust project, add it to your Cargo.toml:
[dependencies]
autogpt = { version = "0.4.5", features = ["gem", "gpt"] }
tokio = { version = "1", features = ["full"] }Select only the feature flags you need to minimize compile times and binary size.
rustup update stable to ensure you have a compatible toolchain.