fal
Fast serverless inference API for image, video, audio and LLM models.
Paid
Fast serverless inference API for image, video, audio and LLM models.
fal is a generative media platform that runs hundreds of hosted models (image, video, audio, LLMs) through a single serverless API with pay-per-use pricing. It also supports deploying your own models on GPUs.
Self-hosted, OpenAI-compatible API for LLMs, speech, images, and agents
Run and deploy open-source AI models with a cloud API.
Run open-source language models locally, with optional cloud for larger ones
Low-cost inference API for open-source LLMs and AI models.
Open-source proxy that gives one API for many LLM providers.