Fine-Tuning on AI Knowledge Base

Fine-Tuning Large Language Models

Mon, 01 Jan 0001 00:00:00 +0000

One-hour introduction to when and how to fine-tune LLMs — covers the decision framework (prompt engineering vs. fine-tuning vs. RAG), data preparation, training process, and evaluation. The fastest way to understand whether fine-tuning is right for your use case before committing engineering resources. Pairs with the Fine-Tuning Across Providers learning path for provider-specific implementation details.

Practical Deep Learning for Coders

Mon, 01 Jan 0001 00:00:00 +0000

Jeremy Howard’s legendary top-down deep learning course — you build working models in lesson 1 and learn theory as needed. This course has produced alumni at Google Brain, OpenAI, Adobe, and Tesla. The teaching philosophy is radical: start with practical results, then go deeper. Covers computer vision, NLP, tabular data, and deployment using PyTorch, fastai, and Hugging Face Transformers. If you learn by doing rather than by theory, this is the best deep learning course available anywhere.

NLP Course

Mon, 01 Jan 0001 00:00:00 +0000

Hugging Face’s comprehensive NLP course — covers tokenizers, transformers, fine-tuning pretrained models, and the Hugging Face ecosystem (Datasets, Tokenizers, Transformers, Accelerate). The definitive open-source-first approach to NLP: everything runs on Hugging Face infrastructure with free GPU access. If you’re building with open models rather than proprietary APIs, start here.

LLM Course

Mon, 01 Jan 0001 00:00:00 +0000

Hugging Face’s newer LLM-specific course covering the full stack: using LLMs, building LLM applications, fine-tuning, and deploying. More focused on the modern LLM workflow than the NLP course. Includes RAG, quantization, and model evaluation. Take this for an open-source-first perspective on the same topics covered by the provider-specific learning paths in this knowledge base.

Fine-Tuning Across Providers

Mon, 01 Jan 0001 00:00:00 +0000

Master fine-tuning across multiple providers — from data preparation to training to deployment. This advanced path covers OpenAI, Mistral, and Together AI’s fine-tuning workflows, with W&B for experiment tracking.

Fine-tuning is a powerful but expensive technique. This path emphasizes the decision framework (when to fine-tune vs alternatives), practical data preparation, and cross-provider comparison of capabilities, costs, and workflows. LoRA and reinforcement fine-tuning expand the toolkit beyond basic supervised fine-tuning.

01 Intro Basics

Mon, 01 Jan 0001 00:00:00 +0000

Learn the basics of fine-tuning LLMs with Mistral AI’s API and open-source tools for optimized performance

02 Prepare Dataset

Mon, 01 Jan 0001 00:00:00 +0000

Learn how to prepare datasets for fine-tuning models across various use cases, from tone to coding and RAG

Agent Tracing

Mon, 01 Jan 0001 00:00:00 +0000

Understand where your agent runs and how tracing enables reinforcement fine-tuning

Available models

Mon, 01 Jan 0001 00:00:00 +0000

See the models you can train with Serverless RL

Azure OpenAI Fine-Tuning

Mon, 01 Jan 0001 00:00:00 +0000

How to Fine-Tune Azure OpenAI models using W&B.

Basics

Mon, 01 Jan 0001 00:00:00 +0000

Understand the reinforcement learning fundamentals behind RFT

Build with Fireworks AI

Mon, 01 Jan 0001 00:00:00 +0000

Fast inference and fine-tuning for open source models

Checkpoints and Resume

Mon, 01 Jan 0001 00:00:00 +0000

Save training progress, resume from failures, and promote checkpoints to deployable models — driven by the recipe.

Classification Finetuning

Mon, 01 Jan 0001 00:00:00 +0000

Classifier Factory

Mon, 01 Jan 0001 00:00:00 +0000

Create and fine-tune custom classification models for intent detection, moderation, sentiment analysis, and more using Mistral’s Classifier Factory

Cleanup and Teardown

Mon, 01 Jan 0001 00:00:00 +0000

Delete trainer jobs and deployments after experiments to avoid leaked resources.

Cookbook Reference

Mon, 01 Jan 0001 00:00:00 +0000

Configuration classes, checkpoint utilities, and gradient accumulation normalization for cookbook recipes.

Cookbook: Distillation

Mon, 01 Jan 0001 00:00:00 +0000

Single-teacher OPD and routed multi-teacher policy distillation with cookbook recipes.

Cookbook: DPO

Mon, 01 Jan 0001 00:00:00 +0000

Direct Preference Optimization with pairwise data using the cookbook recipe.

Cookbook: Reinforcement Learning

Mon, 01 Jan 0001 00:00:00 +0000

Async RL on Fireworks — write a rollout function, the recipe owns the loop (gate, advantage, weight sync, KL/TIS, PPO, checkpoints). Runs async or fully synchronous.

Cookbook: SFT

Mon, 01 Jan 0001 00:00:00 +0000

Supervised fine-tuning via the cookbook’s sft_loop recipe.

Cost Estimator

Mon, 01 Jan 0001 00:00:00 +0000

Estimate and optimize the cost of your RFT training jobs

Create RL Training Job

Mon, 01 Jan 0001 00:00:00 +0000

Create SFT Training Job

Mon, 01 Jan 0001 00:00:00 +0000

Create a new SFT (Supervised Fine-Tuning) training job.

Creating a fine-tuned LoRA

Mon, 01 Jan 0001 00:00:00 +0000

Learn how to create a fine-tuned LoRA to use with W&B Inference.

Data Preparation

Mon, 01 Jan 0001 00:00:00 +0000

Together Fine-tuning API accepts two data formats for training dataset files: text data and tokenized data (in the form of Parquet files). Below, you can learn about different types of those formats and the scenarios in which they can be most useful.

Data Security

Mon, 01 Jan 0001 00:00:00 +0000

How we secure and handle your data for inference and training

Debug SFT tokenization

Mon, 01 Jan 0001 00:00:00 +0000

Download rendered token IDs and loss masks for supervised fine-tuning jobs.

Deploy Finetuned Command Models from AWS Marketplace

Mon, 01 Jan 0001 00:00:00 +0000

This document provides a guide for bringing your own finetuned models to Amazon SageMaker.

Deploy your finetuned model on AWS Marketplace

Mon, 01 Jan 0001 00:00:00 +0000

Learn how to deploy your finetuned model on AWS Marketplace.

Deploying a Fine-tuned Model

Mon, 01 Jan 0001 00:00:00 +0000

Once your fine-tune job completes, you should see your new model in your models dashboard.

Deploying Fine Tuned Models

Mon, 01 Jan 0001 00:00:00 +0000

Deploy one or multiple LoRA models fine tuned on Fireworks using live merge or multi-LoRA

DeploymentManager (Compatibility)

Mon, 01 Jan 0001 00:00:00 +0000

Legacy SDK reference for direct deployment lifecycle and weight-sync management.

DeploymentSampler

Mon, 01 Jan 0001 00:00:00 +0000

Client-side tokenized sampling from inference deployments for training and evaluation.

Direct preference optimization

Mon, 01 Jan 0001 00:00:00 +0000

Fine-tune models for subjective decision-making by comparing model outputs.

Direct Preference Optimization

Mon, 01 Jan 0001 00:00:00 +0000

Distillation

Mon, 01 Jan 0001 00:00:00 +0000

Distillation Trainer

Mon, 01 Jan 0001 00:00:00 +0000

Distributing Training

Mon, 01 Jan 0001 00:00:00 +0000

download the validation and reformat script

Mon, 01 Jan 0001 00:00:00 +0000

Download the reformat_data.py script to validate and reformat datasets for Mistral API fine-tuning

Evaluators

Mon, 01 Jan 0001 00:00:00 +0000

Understand the fundamentals of evaluators and reward functions in reinforcement fine-tuning

Fine Tuning FAQs

Mon, 01 Jan 0001 00:00:00 +0000

Fine-tuning

Mon, 01 Jan 0001 00:00:00 +0000

Fine-tuning models incurs a $2 monthly storage fee per model; see pricing for details

Fine-tuning best practices

Mon, 01 Jan 0001 00:00:00 +0000

Practical tips for getting fine-tuning right — dataset quality vs quantity, avoiding overfitting, hyperparameter selection, and debugging poor results.

Fine-tuning BYOM

Mon, 01 Jan 0001 00:00:00 +0000

Bring Your Own Model: Fine-tune Custom Models from the Hugging Face Hub

Fine-tuning Guide

Mon, 01 Jan 0001 00:00:00 +0000

Learn the basics and best practices of fine-tuning large language models.

Finetuning Agents

Mon, 01 Jan 0001 00:00:00 +0000

Finetuning Cohere Models on AWS Sagemaker

Mon, 01 Jan 0001 00:00:00 +0000

Learn how to finetune one of Cohere’s models on AWS Sagemaker.

Finetuning on Cohere's Platform

Mon, 01 Jan 0001 00:00:00 +0000

An example of finetuning using Cohere’s platform and a financial dataset.

FiretitanServiceClient & TrainingClient

Mon, 01 Jan 0001 00:00:00 +0000

Connect to a trainer endpoint and use the training client for forward/backward passes, optimizer steps, and checkpointing.

Fireworks Agent Overview

Mon, 01 Jan 0001 00:00:00 +0000

Describe what you want, approve the plan and cost, get a deployed fine-tuned model.

Fireworks Agent: Classification

Mon, 01 Jan 0001 00:00:00 +0000

Benchmark base models, fine-tune on labeled data, and pick the best classifier — automatically.

Fireworks Agent: Evaluator Authoring

Mon, 01 Jan 0001 00:00:00 +0000

Have Fireworks Agent generate a reusable evaluator from your dataset — for scoring candidates in an SFT sweep, or for use with Managed RFT.

Fireworks Agent: Preference Learning (DPO/ORPO)

Mon, 01 Jan 0001 00:00:00 +0000

Run preference fine-tuning end-to-end with optional base-model sweep, automatic pair generation, and pairwise evaluation.

Fireworks Agent: Supervised Fine-Tuning

Mon, 01 Jan 0001 00:00:00 +0000

Run end-to-end SFT with Fireworks Agent — dataset inspection, hyperparameter sweep, evaluator-guided selection, and a deployed winner.

FireworksClient

Mon, 01 Jan 0001 00:00:00 +0000

Account-level operations that don’t require a running trainer job.

Function Calling Fine-tuning

Mon, 01 Jan 0001 00:00:00 +0000

Learn how to fine-tune models with function calling capabilities using Together AI.

General Online Logit Distillation (GOLD) Trainer

Mon, 01 Jan 0001 00:00:00 +0000

Generalized Knowledge Distillation Trainer

Mon, 01 Jan 0001 00:00:00 +0000

get data from hugging face

Mon, 01 Jan 0001 00:00:00 +0000

FAQ on data validation, size limits, job creation, and fine-tuning details for Mistral API and mistral-finetune

Get Training Job

Mon, 01 Jan 0001 00:00:00 +0000

Get a training job by ID.

Get Training Job Events

Mon, 01 Jan 0001 00:00:00 +0000

Get events for a training job.

Glossary

Mon, 01 Jan 0001 00:00:00 +0000

Glossary of key AI and LLM terms, including LLMs, text generation, tokens, MoE, RAG, fine-tuning, function calling, embeddings, and temperature

GPU Clusters Overview

Mon, 01 Jan 0001 00:00:00 +0000

High-performance GPU clusters for training, fine-tuning, and large-scale AI workloads

Graders

Mon, 01 Jan 0001 00:00:00 +0000

Learn about graders used for evals and fine-tuning.

How to use Serverless SFT

Mon, 01 Jan 0001 00:00:00 +0000

Hugging Face Accelerate

Mon, 01 Jan 0001 00:00:00 +0000

Training and inference at scale made simple, efficient and adaptable

Incremental Snapshots (ARC2)

Mon, 01 Jan 0001 00:00:00 +0000

Build ARC2 incremental checkpoints, use per-file hints, and signal delta hot-loads for BYOT RL rollout integrations.

Introduction

Mon, 01 Jan 0001 00:00:00 +0000

Fireworks Training API — custom training loops with full Python control over objectives, while Fireworks handles distributed GPU infrastructure.

Ledger & Debugging for RL Rollouts

Mon, 01 Jan 0001 00:00:00 +0000

Inspect snapshot history, reset the ledger, and understand how in-flight requests behave during a weight swap.

Log distributed training experiments

Mon, 01 Jan 0001 00:00:00 +0000

Use W&B to log distributed training experiments with multiple GPUs.

LoRA Fine-Tuning and Inference

Mon, 01 Jan 0001 00:00:00 +0000

Fine-tune and run inference for a model with LoRA adapters

Loss Functions

Mon, 01 Jan 0001 00:00:00 +0000

Built-in loss functions and custom objectives via forward_backward_custom.

Managed Fine-Tuning Overview

Mon, 01 Jan 0001 00:00:00 +0000

Fine-tune models with Fireworks-managed infrastructure — no custom code required.

Model customization

Mon, 01 Jan 0001 00:00:00 +0000

Learn how to customize LLMs for your application with system prompts, fine-tuning, and moderation layers

Model optimization

Mon, 01 Jan 0001 00:00:00 +0000

Ensure quality model outputs with evals and fine-tuning in the OpenAI platform.

Monitor Training

Mon, 01 Jan 0001 00:00:00 +0000

Track RFT job progress and diagnose issues in real-time

OpenAI Fine-Tuning

Mon, 01 Jan 0001 00:00:00 +0000

How to Fine-Tune OpenAI models using W&B.

openapi

Mon, 01 Jan 0001 00:00:00 +0000

OpenEnv Integration for Training LLMs with Environments

Mon, 01 Jan 0001 00:00:00 +0000

OpenReward Integration for Training LLMs with Environments

Mon, 01 Jan 0001 00:00:00 +0000

Optimizing LLM Accuracy

Mon, 01 Jan 0001 00:00:00 +0000

Learn strategies to enhance the accuracy of large language models using techniques like prompt engineering, retrieval-augmented generation, and fine-tuning.

Orthogonal Finetuning (OFT and BOFT)

Mon, 01 Jan 0001 00:00:00 +0000

Overview

Mon, 01 Jan 0001 00:00:00 +0000

Train models using reinforcement learning in minutes

Overview

Mon, 01 Jan 0001 00:00:00 +0000

Welcome to Together AI’s docs! Together makes it easy to run, finetune, and train open source AI models with transparency and privacy.

Parameter Tuning

Mon, 01 Jan 0001 00:00:00 +0000

Learn how training parameters affect model behavior and outcomes

Post-Training Toolkit Integration

Mon, 01 Jan 0001 00:00:00 +0000

Preference Fine-Tuning

Mon, 01 Jan 0001 00:00:00 +0000

Learn how to use preference fine-tuning on Together Fine-Tuning Platform

Prerequisites

Mon, 01 Jan 0001 00:00:00 +0000

Set up your environment to use W&B Training

Price comparison vs Tinker

Mon, 01 Jan 0001 00:00:00 +0000

Estimate the cost of multi-turn agentic RL rollouts on Fireworks compared to Tinker’s per-token pricing

Pricing

Mon, 01 Jan 0001 00:00:00 +0000

Fine-tuning pricing at Together AI is based on the total number of tokens processed during your job.

Quickstart

Mon, 01 Jan 0001 00:00:00 +0000

Get a custom training loop running in minutes with the Fireworks Training API.

Reasoning Fine-tuning

Mon, 01 Jan 0001 00:00:00 +0000

Learn how to fine-tune reasoning models with chain-of-thought data using Together AI.

Reinforcement fine-tuning

Mon, 01 Jan 0001 00:00:00 +0000

Train models using reward-based signals for expert-level domain performance — going beyond supervised fine-tuning when you can grade quality but can’t easily provide gold outputs.

Reinforcement fine-tuning use cases

Mon, 01 Jan 0001 00:00:00 +0000

Practical use cases and best practices for reinforcement fine-tuning (RFT) — when graded rewards outperform simple input/output pairs for training signal.

Remote Agent Quickstart

Mon, 01 Jan 0001 00:00:00 +0000

Train an SVG drawing agent running in a remote environment

Remote Environment Setup

Mon, 01 Jan 0001 00:00:00 +0000

Implement the /init endpoint to run evaluations in your infrastructure

RFT parameters reference

Mon, 01 Jan 0001 00:00:00 +0000

Checkpoint, resume, and GRPO metrics fields for reinforcement fine-tuning recipes.

RL Rollouts with Your Own Trainer

Mon, 01 Jan 0001 00:00:00 +0000

Integrate an external RL trainer with Fireworks inference: hot-load new checkpoints from your bucket and run rollouts via the OpenAI-compatible API.

Saving and Loading

Mon, 01 Jan 0001 00:00:00 +0000

SDK-level reference for checkpoint save, load, weight sync, and promotion.

Secure Training (BYOB)

Mon, 01 Jan 0001 00:00:00 +0000

Fine-tune models while keeping sensitive data and components under your control

Single-Turn Training Quickstart

Mon, 01 Jan 0001 00:00:00 +0000

Train a model to be an expert at answering GSM8K math questions

Speeding Up Training

Mon, 01 Jan 0001 00:00:00 +0000

Supervised Fine Tuning - Text

Mon, 01 Jan 0001 00:00:00 +0000

Supervised Fine Tuning - Vision

Mon, 01 Jan 0001 00:00:00 +0000

Learn how to fine-tune vision-language models on Fireworks AI with image and text datasets

Supervised fine-tuning

Mon, 01 Jan 0001 00:00:00 +0000

Fine-tune OpenAI models with input/output pairs — covering data preparation, training configuration, validation, and deployment of customized models.

Supported Models

Mon, 01 Jan 0001 00:00:00 +0000

A list of all the models available for fine-tuning.

Text & Vision Fine-tuning

Mon, 01 Jan 0001 00:00:00 +0000

Fine-tune Mistral’s text and vision models with custom datasets in JSONL format for domain-specific or conversational improvements

The Cookbook

Mon, 01 Jan 0001 00:00:00 +0000

Ready-to-run training recipes for GRPO, DPO, SFT, and distillation built on top of the Training API.

Together AI

Mon, 01 Jan 0001 00:00:00 +0000

Track and evaluate Together AI’s open source LLMs using Weave’s OpenAI SDK compatibility for seamless integration with model calls, fine-tuning workflows, and hosted models.

Together AI Skills

Mon, 01 Jan 0001 00:00:00 +0000

Give your AI coding agent deep knowledge of the Together AI platform with ready-made skills for inference, training, images, video, audio, and infrastructure.

TrainerJobManager (Compatibility)

Mon, 01 Jan 0001 00:00:00 +0000

Legacy SDK reference for service-mode trainer job lifecycle management.

Training

Mon, 01 Jan 0001 00:00:00 +0000

Learn how to train your CrewAI agents by giving them feedback early on and get consistent results.

Training and Sampling

Mon, 01 Jan 0001 00:00:00 +0000

End-to-end SDK walkthrough: bootstrap resources, train, checkpoint, and sample through a serving deployment.

Training customization

Mon, 01 Jan 0001 00:00:00 +0000

Training Guide: UI

Mon, 01 Jan 0001 00:00:00 +0000

Launch RFT jobs using the Fireworks dashboard

Training Overview

Mon, 01 Jan 0001 00:00:00 +0000

Launch RFT jobs using the eval-protocol CLI

Training Overview

Mon, 01 Jan 0001 00:00:00 +0000

This overview frames Fireworks’ three fine-tuning paths — the autonomous Agent, semi-managed Managed Fine-Tuning, and the custom Training API — so it matters as the decision page before you commit compute. The key heuristic it offers is to reach for supervised fine-tuning when you have more than about a thousand quality labeled examples, and to switch to reinforcement fine-tuning for smaller datasets or reasoning-heavy tasks where ground-truth labels do not exist. A common mistake is defaulting to SFT on too little data. This is the Fireworks counterpart to Together AI’s fine-tuning flow; read the quickstart first if you are new to the platform.

Training Prerequisites & Validation

Mon, 01 Jan 0001 00:00:00 +0000

Requirements, validation checks, and common issues when launching RFT jobs

Training Shapes

Mon, 01 Jan 0001 00:00:00 +0000

Pre-configured GPU and model training profiles that simplify distributed training setup.

Training with Jobs

Mon, 01 Jan 0001 00:00:00 +0000

Understanding LoRA performance

Mon, 01 Jan 0001 00:00:00 +0000

Understand the performance impact of LoRA fine-tuning, optimization strategies, and deployment considerations.

Upload a Model

Mon, 01 Jan 0001 00:00:00 +0000

Run inference on your custom or fine-tuned models

Usage information and limits

Mon, 01 Jan 0001 00:00:00 +0000

Understand pricing, usage limits, and account restrictions for W&B Serverless RL

Use Fireworks Agent with Claude Code, Cursor, Codex, and other coding agents

Mon, 01 Jan 0001 00:00:00 +0000

Install the Fireworks Agent skill file once and drive end-to-end fine-tuning from your coding agent.

Use model after training

Mon, 01 Jan 0001 00:00:00 +0000

Use Serverless LoRA Inference

Mon, 01 Jan 0001 00:00:00 +0000

Bring your own custom LoRA for serving fine-tuned models on W&B Inference.

Use Weave with W&B training runs

Mon, 01 Jan 0001 00:00:00 +0000

Integrate Weave traces with W&B training runs to gain deep visibility into model behavior during training, capturing function execution details and diagnostics alongside traditional ML metrics in customizable workspace dashboards.

Use your trained models

Mon, 01 Jan 0001 00:00:00 +0000

Make inference requests to the models you’ve trained

Using Secrets

Mon, 01 Jan 0001 00:00:00 +0000

Learn how to create secrets that can be utilized within your reward function.

Verifiers

Mon, 01 Jan 0001 00:00:00 +0000

Track and debug Verifiers RL environments and LLM agent training with Weave, capturing multi-round conversations, evaluation rollouts, and model performance metrics for comprehensive observability of reinforcement learning workflows.

Vision fine-tuning

Mon, 01 Jan 0001 00:00:00 +0000

Fine-tune models for better image understanding.

Vision Inputs

Mon, 01 Jan 0001 00:00:00 +0000

Fine-tune vision-language models (VLMs) with the Training API using multimodal chat data containing images and text.

Vision-Language Fine-tuning

Mon, 01 Jan 0001 00:00:00 +0000

Learn how to fine-tune Vision-Language Models (VLMs) on image+text data using Together AI.

W&B Mobile App (iOS)

Mon, 01 Jan 0001 00:00:00 +0000

Track training runs, view line plots, and explore your W&B Models projects from your iPhone or iPad.

W&B Training

Mon, 01 Jan 0001 00:00:00 +0000

Warm Start from Fine-Tuned Models

Mon, 01 Jan 0001 00:00:00 +0000

Continue training from a previously fine-tuned model with RFT

Weight sync

Mon, 01 Jan 0001 00:00:00 +0000

How a trainer’s updated weights reach the serving deployment during RL training.

Weighted Training

Mon, 01 Jan 0001 00:00:00 +0000

Control which samples have greater influence during RFT training

WeightSyncer (Legacy)

Mon, 01 Jan 0001 00:00:00 +0000

Backward-compatibility reference for the old standalone checkpoint-then-sync helper.

Welcome to LLM University!

Mon, 01 Jan 0001 00:00:00 +0000

LLM University (LLMU) offers in-depth, practical NLP and LLM training. Ideal for all skill levels. Learn, build, and deploy Language AI with Cohere.

Workspaces

Mon, 01 Jan 0001 00:00:00 +0000

La Plateforme workspaces enable team collaboration, access control, and shared fine-tuned models.’ (99 characters)