Cookbooks

Worked examples and application case studies for training, retrieval, and coding agents on Fireworks AI.

Explore notebooks and application case studies for building with Fireworks AI.

​

Training

VLM Training + Evals

Beat frontier closed-source models for product catalog cleansing with vision-language model training.

Techniques: Supervised Fine-Tuning (SFT)

Results: 48% increase in quality from base model

​

Embeddings & RAG

Two-Stage Retrieval with Voyage Embeddings and Reranking

Build a retrieval pipeline that recalls candidates with vector search, then reorders them with a reranker for precision.

Techniques: Voyage embeddings, MongoDB $vectorSearch, Voyage ReRank 2.5

Models: Voyage AI embedders and rerankers on dedicated deployments

​

Coding agents

Advisor: a second-opinion code reviewer

Add a critique-only frontier reviewer to a coding agent running on Fireworks.