Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.veri.studio/llms.txt

Use this file to discover all available pages before exploring further.

​
Overview

Veri can be used through three developer entry points:
  • the REST API
  • the Python SDK
  • the CLI
They all map to the same core workflow: create datasets, upload reward functions, submit training jobs, poll status, and retrieve final checkpoints.

​
Choose Your Interface

​
REST API

Use the REST API if you want full control over requests, integration with your own services, or language-agnostic access. Best for:
  • backend integrations
  • direct HTTP workflows
  • custom automation outside Python
Start here:

​
Python SDK

Use the Python SDK if you want the fastest programmatic path in Python. The current SDK supports:
  • dataset upload, connect, validate, list, get
  • reward function upload, list, get
  • training job create, get, list, cancel
  • checkpoint download helpers
  • polling with job.wait()
Start here:

​
CLI

Use the CLI if you want a lightweight local workflow for logging in, storing credentials, and bootstrapping SDK usage. The current CLI supports:
  • veri login
  • veri whoami
  • veri logout
Start here:

​
Typical Flow

# 1. Log in locally
veri login

# 2. Use the SDK or raw API calls
# 3. Submit a training job
# 4. Poll job status
# 5. Download or deploy the final checkpoint

​
Current Notes

  • The API surface is moving faster than the polished public auth story.
  • The Python SDK already mirrors the main control-plane resources.
  • The CLI is intentionally minimal and focused on credential management.
If you are unsure where to start, the best path today is:
  1. CLI for local setup
  2. Python SDK for scripting
  3. API Reference for exact request shapes