Test1234

FIREWORKS AI CLOUD

Build. Train. Scale.

Open-source AI models at blazing speed, optimized for your use case, scaled globally with the Fireworks Inference Cloud

Get Started

Talk to our team

test

Mux test

test mux section 1

test--mux section 2 - no auto-expand

Test mux video - hide side margins false

test iframe embed

Test iframe embed - hide side margins true

50%

Higher throughput per GPU

Test section with video - autoplay on click?

test

Rive

TEST

Animated SVG Image File

Question 1

Answer 1

Question 2

Answer 2

Question with a long text that may overflow the container -- this is a test

Lorem ipsum dolor sit amet, consectetur adipiscing elit. In eu suscipit diam. Vestibulum luctus, magna feugiat dapibus feugiat, lorem tortor auctor turpis, quis aliquam magna massa sit amet purus. Curabitur iaculis odio tellus, vitae aliquet magna hendrerit at. Donec sem justo, feugiat ut commodo non, rhoncus in ligula. Donec sed elementum mi, non lobortis neque. Pellentesque aliquet, neque et pellentesque rutrum, ex lectus interdum dui, sit amet rhoncus ex ipsum vehicula sem. Nam tincidunt leo sed augue dictum, vitae malesuada lectus posuere. Aliquam nisl ipsum, luctus sed erat ut, molestie congue lacus. Praesent placerat tristique odio, et rhoncus elit finibus at. Phasellus pellentesque libero pulvinar nunc molestie, sed gravida tortor fringilla. Donec non laoreet elit. Praesent tortor lectus, fermentum non accumsan quis, ultricies id eros. Curabitur blandit ante lorem, quis tincidunt orci rhoncus et. Maecenas condimentum egestas arcu, id condimentum lacus feugiat eu. Vestibulum sollicitudin nisl arcu, vitae fermentum justo bibendum a. In hac habitasse platea dictumst.

eyebrow

FAQ

FAQ Component

BUTTON

Question 1

Answer 1

Question 2

Answer 2

Question with a long text that may overflow the container -- this is a test

Lorem ipsum dolor sit amet, consectetur adipiscing elit. In eu suscipit diam. Vestibulum luctus, magna feugiat dapibus feugiat, lorem tortor auctor turpis, quis aliquam magna massa sit amet purus. Curabitur iaculis odio tellus, vitae aliquet magna hendrerit at. Donec sem justo, feugiat ut commodo non, rhoncus in ligula. Donec sed elementum mi, non lobortis neque. Pellentesque aliquet, neque et pellentesque rutrum, ex lectus interdum dui, sit amet rhoncus ex ipsum vehicula sem. Nam tincidunt leo sed augue dictum, vitae malesuada lectus posuere. Aliquam nisl ipsum, luctus sed erat ut, molestie congue lacus. Praesent placerat tristique odio, et rhoncus elit finibus at. Phasellus pellentesque libero pulvinar nunc molestie, sed gravida tortor fringilla. Donec non laoreet elit. Praesent tortor lectus, fermentum non accumsan quis, ultricies id eros. Curabitur blandit ante lorem, quis tincidunt orci rhoncus et. Maecenas condimentum egestas arcu, id condimentum lacus feugiat eu. Vestibulum sollicitudin nisl arcu, vitae fermentum justo bibendum a. In hac habitasse platea dictumst.

FAQs

Item 1

Answer 1

•a
•b
•c
•d
•e

Item 2

test of richtext

–dashed

checked

numbered

•
bulleted
- •level 2

This partnership isn’t just about adding new models to Microsoft Foundry, it’s about giving teams better tools to build, run, and scale impactful AI use cases. Coming soon, our roadmap brings fine‑tuning capabilities for Fireworks AI models into Microsoft Foundry to deliver the best end‑to‑end destination for open‑source model customization & deployment. We’re looking forward to seeing how developers and enterprises use Fireworks on Foundry to power the next generation of intelligent applications.

Test

Llama-70B GRPO with tool rewards

Distributed GRPO on a 70B base, 256 rollout workers, model-graded reward. Used by Genspark.
★ 312 · forks 47

SFT for code-edit completion

Two-epoch. SFT on edit pairs with prefix masking. The recipe Cursor open-sourced.
★ 1,840 · forks 261

DPO with model-graded preferences

Sample N completions per prompt, judge with a stronger model, train DPO end-to-end.
★ 720 · forks 94

eyebrow

Fullwidth video switchback

test

eyebrow

Normal video switchback

test

Train where you want. Roll out on Fireworks.

Whether you want a fully managed workflow or a Tinker-compatible setup, Fireworks can power the rollout layer. Bring your trainer — Fireworks handles rollout inference, policy updates, and fleet orchestration.

Train where you want. Roll out on Fireworks.

Whether you want a fully managed workflow or a Tinker-compatible setup, Fireworks can power the rollout layer. Bring your trainer — Fireworks handles rollout inference, policy updates, and fleet orchestration.

test

Training API · Beta