Cloud platform to run open-source AI models via API
Replicate makes it easy to run machine learning models in the cloud with a simple API. Access thousands of open-source models—Stable Diffusion, Llama, Flux, Whisper—or deploy your own. Pay-per-second billing. Integrated directly into Hugging Face Hub in January 2025.
4 ready-to-use prompts — click any card to copy, or paste into the writer above.
Using the Replicate API with Python, run the SDXL model to generate a product photograph. Input: 'a minimalist ceramic coffee mug on a wooden table, morning light, lifestyle photography'. Set: width=1024, height=1024, num_inference_steps=30, guidance_scale=7.5. Print the output URL.
Use Replicate to upscale a low-resolution video file using the Real-ESRGAN model. Take the input video URL from a command-line argument, poll for completion every 5 seconds, download the result when done, and save with '_upscaled' suffix.
Call the Whisper model on Replicate to transcribe a podcast episode audio file. Return the full transcript with timestamps every 30 seconds, detect the language automatically, and format the output as an SRT subtitle file.
Write a Python script using the Replicate API that takes a folder of product images, runs background removal on each using the rembg model, saves the transparent PNG results to an output folder, and handles rate limiting gracefully with a queue.
