Fireworks AI
Fireworks AI develops infrastructure for running and adapting AI models. Its platform combines hosted inference with tools for training and customizing models on business data. The company serves developers and organizations deploying production AI applications, with model access and compute governed by the chosen service.
- Status
- Active
- Checked 29 Oct 2026
- Tools listed
- 1
Tools from Fireworks AI
- Fireworks AI Inference Platform
Run, ensemble and fine-tune open AI models through one API platform.
Apps- Web ↗
Create an account, then use the dashboard to manage API keys, usage, deployments and the playground. Existing users sign in at the login URL.
Developer access- API ↗
Create an API key in the dashboard or with firectl, then call the REST API. The API is OpenAI-compatible, with a base URL of https://api.fireworks.ai/inference/v1.
- Command line (firectl) ↗
firectl installs via Homebrew on macOS, or as a downloaded binary on macOS, Linux and Windows. Run `firectl signin` to authenticate in the browser. It is a command-line tool, not a GUI app.
- SDK / library (Python (fireworks-ai)) ↗
Install with `pip install --upgrade fireworks-ai` and set FIREWORKS_API_KEY.
Verified options shown; other ways to use this tool are still being checked.
Coding · Sign up · Checked
Usage-basedPer token, per training token, or per GPU hour - Web ↗
When Fireworks AI launches another product, it gets its own record here.
@FireworksAI_HQ on X
We read everything Fireworks AI posts on X and summarize what’s worth your time: launches, new features and price changes. The noise stays out.