Tutorials

Step-by-step walkthroughs with screenshots of Arena. Each tutorial shows where to click and what you should see on screen.

Home links to these tutorials under Written tutorials.

Area

Start here

Environments

Environments

Datasets

Datasets

Experiments

Experiments

Agents

Agents

Account

Account

LLM data

Tutorial

What you will do

Upload a reasoning dataset

Create a Language Reasoning dataset (Question / Answer)

Upload a preference dataset

Create a Language Preference dataset (Prompt / Chosen / Rejected)

Upload an SFT dataset

Create a Language SFT dataset (Prompt / Target)

Upload a language-based environment

Add a language-based gym for LLM simulation

LLM training

Tutorial

What you will do

Create a training project

Create a project with Advanced Training enabled and open Datasets

Run an SFT experiment

SFT dataset + algorithm SFT through the wizard

Run a Preference experiment

Preference dataset + DPO through the wizard and start training

Run a Reasoning experiment

Reasoning dataset + CISPO, including prompt and reward validation

Run a Multi-turn LLM experiment

Language-based gym + LLMPPO with advantage_granularity turn

View results and checkpoints

Open metrics and checkpoint rows after a run

Agents

Tutorial

What you will do

Deploy and invoke an agent

Deploy from a checkpoint and send a test prompt (UI, CLI, Python)

Account

Tutorial

What you will do

Billing and balance

Read usage and the balance (org Admin)

Before you start

  • Sign in to Arena with an account that can create projects in your organization.

  • Training runs spend balance. The screenshots use sample org Acme Lab with stable demo data.