Skip to main content

​
What is the Cookbook?

The Fireworks Cookbook is a collection of training recipes and utilities built on top of the Training API. It provides config-driven training loops that handle trainer provisioning, data loading, tokenization, gradient accumulation, checkpointing, and cleanup automatically. The cookbook is optional — everything it does can be done with the API directly. Use the cookbook when you want a working training loop quickly; use the API when you need full control.

​
Installation

git clone https://github.com/fw-ai/cookbook.git
cd cookbook/training && pip install -e .
Set your credentials: 
export FIREWORKS_API_KEY="your-api-key"

​
Available recipes

RecipeModuleUse case
RLtraining.recipes.rl_loopOn-policy and off-policy reinforcement learning with GRPO, importance sampling, DAPO, DRO, GSPO, and CISPO
DPOtraining.recipes.dpo_loopDirect preference optimization from chosen/rejected pairs
SFTtraining.recipes.sft_loopSupervised fine-tuning with cross-entropy loss
ORPOtraining.recipes.orpo_loopOdds ratio preference optimization
Each recipe follows the same pattern: import Config and main, set your config, and call main(cfg). All launch examples below use infra=InfraConfig(training_shape_id=...). For cookbook users, that training shape ID is usually the only shape-specific input you need to set. If you want field-level details about what a training shape controls and what stays configurable, see the API reference pages linked from Training Shapes.

​
Quick example: SFT

from training.recipes.sft_loop import Config, main
from training.utils import InfraConfig

cfg = Config(
    log_path="./sft_quickstart",
    base_model="accounts/fireworks/models/qwen3-8b",
    dataset="/path/to/training_data.jsonl",
    tokenizer_model="Qwen/Qwen3-8B",
    max_seq_len=4096,
    epochs=1,
    batch_size=4,
    infra=InfraConfig(
        training_shape_id="accounts/fireworks/trainingShapes/qwen3-8b-128k-h200",
    ),
)

main(cfg)

​
Quick example: GRPO

from training.recipes.rl_loop import Config, main
from training.utils import DeployConfig, InfraConfig, WeightSyncConfig

cfg = Config(
    log_path="./grpo_quickstart",
    base_model="accounts/fireworks/models/qwen3-8b",
    dataset="/path/to/prompts.jsonl",
    max_rows=100,
    infra=InfraConfig(
        training_shape_id="accounts/fireworks/trainingShapes/qwen3-8b-128k-h200",
    ),
    deployment=DeployConfig(
        deployment_id="grpo-serving",
        tokenizer_model="Qwen/Qwen3-8B",
    ),
    weight_sync=WeightSyncConfig(weight_sync_interval=1),
)

main(cfg)

​
W&B logging

All cookbook recipes accept a WandBConfig to stream metrics to Weights & Biases: 
from training.utils import WandBConfig

cfg = Config(
    # ... same config as above ...
    wandb=WandBConfig(
        entity="my-team",
        project="grpo-experiment",
        run_name="qwen3-8b-sft-v1",  # optional, auto-generated if omitted
    ),
)

main(cfg)

​
Vision-language model support

All cookbook recipes support VLM fine-tuning. Use a VLM training shape and tokenizer, and provide multimodal datasets with image_url content. See Vision Inputs for dataset format and examples.

​
Next steps