Endpoints

Pay only for request execution time.

Launch your product today.

Autoscale
Get started quickly
Speech Recognition
Generative Art
3,610,816,821
requests since launch

Large Language Models

Llama2 13B
$0.00185 / 1000 tokens
48 GB VRAM
Llama2 7B
$0.00075 / 1000 tokens
24 GB VRAM
Pygmalion 6B
$0.00055 / second
48 GB VRAM
LLM Prompt
RunPod API KeyFind my API Key
GPU Type
Max Tokens
For a complete list of paramaters, check out our API Documentation

Speech Recognition

Whisper
$0.00025 / second
3 minutes of audio in 30s
Choose between various models
Faster Whisper
$0.00025 / second
3 minutes of audio in 11s
2-4x Faster than vanilla Whisper

Text to Image

Anything V3
v1.5
$0.00025 / second
24 GB VRAM (768x768 max)
3.4s for 512x512 25 steps
5,000 images for $4.25
Anything V4
v1.5
$0.00025 / second
24 GB VRAM (768x768 max)
3.4s for 512x512 25 steps
5,000 images for $4.25
DreamBooth
v1.5
$0.001 / second
80 GB VRAM
4m training time for 1000 steps
100s of images in less than 10m
Openjourney
v1.5
$0.00025 / second
24 GB VRAM (768x768 max)
3.4s for 512x512 25 steps
5,000 images for $4.25
Stable Diffusion
v1.5
$0.00025 / second
24 GB VRAM (768x768 max)
3.4s for 512x512 25 steps
5,000 images for $4.25
Stable Diffusion
v2
$0.00025 / second
24 GB VRAM (768x768 max)
3.4s for 512x512 25 steps
5,000 images for $4.25
Kandinsky
v2.1
$0.00025 / second
24 GB VRAM (768x768 max)
Supports multi-language
Better coherence than SD
Get an API Key
Data Security & Usage
Endpoints temporarily save data to fulfill requests and allow status checks within 30 mins of completion.