> ## Documentation Index
> Fetch the complete documentation index at: https://fireworks.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Training cost estimator

> Estimate Managed Training, compare Fireworks Serverless with Dedicated, or compare Fireworks Dedicated with Tinker

> [!NOTE]
>
> <ul>
>   <li>
>     Managed and Serverless are [priced per
>     token](https://fireworks.ai/pricing#training-pricing).
>   </li>
>
>   <li>
>     Dedicated is [priced per allocated GPU
>     hour](https://fireworks.ai/pricing#on-demand-pricing).
>   </li>
> </ul>

Training cost estimator

## Configure once. Compare every supported path.

Keep the method, model, workload, and assumptions visible while
estimating Managed and comparing Fireworks Serverless vs Dedicated or
Fireworks Dedicated vs Tinker.

Training method

> [!WARNING]
>
> <ul>
>   <li>Planning estimates are not quotes.</li>
>
>   <li>
>     Managed estimates can be low when rendered-token inputs omit multi-turn
>     unrolling. Dedicated compute floors can be low when sequences pack poorly.
>   </li>
>
>   <li>
>     Dedicated shows a saturated compute floor. Allocated GPU time spent
>     initializing the model, writing checkpoints, or idle can make real jobs
>     cost more. Queue wait and pre-allocation provisioning are not billed.
>   </li>
> </ul>

## Prepare inputs with the Training Skill

The [Fireworks Training Skill](https://fireworks.ai/docs/fine-tuning/agent/use-with-coding-agents.md)
prepares inputs and calculates Managed and Serverless estimates from published
rates. It can inspect a local dataset or an existing job and identify
assumptions. It does not calculate Dedicated numbers. Use this page for
Dedicated planning.

The Skill does not launch a job or authorize spend while estimating. Training
still requires the Skill's complete final plan and your explicit confirmation.

## Reinforcement learning

RL cost varies with rollout shape, concurrency, reward or verifier design,
training method, and evaluation workload. [Contact the Training
team](https://fireworks.ai/contact-training) for a tailored estimate.

To compare the rollout inference cost of multi-turn agentic RL, use the separate
[rollout cost comparison](https://fireworks.ai/docs/fine-tuning/multi-turn-cost-comparison.md). That page
does not estimate SFT or DPO training cost.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.