(function() {
    const scriptId = 'hs-script-loader';
    if (document.getElementById(scriptId) !== null) {
        return;
    }
    
    // Create the script element
    const script = document.createElement('script');
    
    // Set the attributes
    script.type = 'text/javascript';
    script.id = scriptId;
    script.async = true;
    script.defer = true;
    script.src = 'https://js.hs-scripts.com/21944583.js';
    
    // Add the script to the body
    document.body.appendChild(script);
})();

Fine-tune a model to play Countdown using reinforcement learning

GRPO for Countdown

Predibase

Learn about Predibase and how to get started

Introduction

Quickstart

Overview

Learn how to query models using the Python SDK or REST API

Querying Models

Learn how to use custom fine-tuned adapters with Predibase models

Fine-tuned Adapters

Process inference requests in bulk at a lower cost

Batch Inference

Models available for fine-tuning in Predibase

Supported Models

Parameter efficient fine-tuning with adapters

Adapters

Fine-tuning with very large datasets in Predibase

Distributed Training

Evaluating the performance of your fine-tuned adapters

Evaluation

Understanding and configuring hyperparameters for fine-tuning

Hyperparameter Tuning

Learn About User Roles and Access Control in Predibase

Roles & Permissions

Integrate Predibase with your favorite ML tools

Integrations

Learn how to fine-tune Mistral-7B for specialized customer support tasks

LoRA Land for Customer Support

Toxic Comment Classifier

Learn how to combine a recommender system with LLM generation for personalized outreach

Recommender System → LLM Generation

Build an end-to-end RAG system with Predibase and LlamaIndex

Retrieval-Augmented Generation

Learn About Trial Credits, Usage Monitoring, and Billing Management

Usage and Billing

Frequently Asked Questions

How to install and set up the Predibase Python SDK

Installation

SDK operations for managing model deployments

Deployments

SDK operations for prompting your deployed models

SDK operations for managing datasets for fine-tuning

Datasets

SDK operations for managing adapter repositories

Repositories

SDK operations for managing and fine-tuning adapters

OpenAI Chat Completions v1 compatible endpoint

Create Chat Completion

OpenAI Completions v1 compatible endpoint

Create Completion

Generate

Generate a stream of tokens using Server-Sent Events

Generate Stream

Health Check

Info

Metrics

Deep dive into parameter efficient fine-tuning with adapters

Guidelines and best practices for preparing data and fine-tuning models

Dataset Preparation

Guidelines and best practices for generating and using synthetic data effectively

Synthetic Data

Guide to reinforcement learning with GRPO

Reinforcement Learning

Learn how to effectively tune hyperparameters for optimal model performance

Welcome to the home of your new documentation

Start building awesome documentation in under 5 minutes

Learn about Predibase's model serving options and how to get started

Deploy dedicated serverless model instances for production use cases

Private Deployments

Quickly experiment with pre-deployed models for development and testing

Shared Endpoints

Learn how to use Predibase's REST API for model inference

REST API

Explore our catalog of supported language models for text generation

Language Models

Explore our catalog of supported vision models for image understanding and generation

Vision Language Models

Explore our catalog of supported embedding models for text embeddings and similarity search

Embedding Models

Learn how to deploy and use custom models from Hugging Face Hub

Custom Base Models

Batch Inference (BETA)

Enforce JSON schema validation on model responses

Structured Output

Easily migrate your OpenAI applications to Predibase

Migrate from OpenAI

Preparing and uploading datasets for fine-tuning

Different types of tasks supported for fine-tuning

Fine-tuning Tasks

Fine-tune models with function calling capabilities

Function Calling

Optimize model behavior using Group Relative Policy Optimization (GRPO)

Reinforcement Fine-tuning

Large Training Jobs

Learn about chat templates and how to use them for fine-tuning

Chat Templates

Users

Reserved Capacity

VPC Provisioning

Generate a stream of token using Server-Sent Events

Getting Started

Inference

Fine-Tuning

Account

Integrations

Examples

Resources

GRPO for Countdown