Get started with distil labs

distil labs is the post-training platform where engineers build high-performance models and continually improve them using production data.

Quickstart

Install the CLI:

curl -fsSL https://cli-assets.distillabs.ai/install.sh | sh

Install the skill for your coding agent, the recommended way to build your first model, and create your account. distil signup opens a browser page where you create the account and sign in:

distil skill install
distil signup

Then start your agent:

Claude CodeCodexCursorAny other agent

claude "Let's get started with distillabs.ai/onboarding.md"

codex "Let's get started with distillabs.ai/onboarding.md"

Let's get started with distillabs.ai/onboarding.md

Let's get started with distillabs.ai/onboarding.md

Next steps

• Account and authentication - create an account and sign in
• Ways to use distil labs - the CLI, the agent skill, the REST API and the dashboard
• Building a model - the loop end to end, and where to enter it
• Task selection - pick the task type first, it sets everything else
• Collect traces with an endpoint - the default: start from the traffic you serve today
• Trace inputs - start from logs you already hold
• Seed datasets - start from a labeled dataset
• Model iterations - improve a trained model, iteration by iteration
• CLI reference - every command
• API reference - the same platform over HTTP