Anyscale
Paid
The managed platform from the creators of Ray for scaling AI training and inference.
All tools
Fireworks AI
A fast inference platform for open models, with tools to tune and deploy them.
About Fireworks AI
Fireworks AI serves open-source and custom models with a focus on speed and production reliability. You pick a model, call the API, and get quick responses without handling GPUs yourself.
It also supports fine-tuning, so you can adapt a model to your data and deploy it on the same platform. Features like structured outputs and function calling make it practical for building agents and compound systems.
Pricing is usage based, with a free credit to start. It is a good fit for companies that want open models running at scale.
Standout features
- Fast open-model inference
- Fine-tuning
- Function calling
- Structured outputs
- Dedicated deployments
- OpenAI-compatible API
Tags
More in LLM Platforms
View all →Comet
Freemium
Experiment tracking and the open-source Opik platform for evaluating LLM apps.
Modal
Freemium
Serverless cloud for running AI workloads, with GPUs available in seconds.
Replicate
Paid
Run open-source and custom AI models in the cloud with a few lines of code.
MLflow
Free
The open-source standard for managing the machine learning lifecycle, now with GenAI tracing.
Weights and Biases
Freemium
Tools for tracking experiments, managing models and evaluating LLM apps.
