Every AI API, priced in the same unit.

Chat, reasoning, coding, agents, and text generation.

$ / 1M out tokens

Turn text or images into vectors for search, RAG and clustering.

$ / 1M tokens

Re-order search results by true relevance.

$ / 1K searches

Dedicated text and document translation APIs.

$ / 1M chars

Models that read photos, screenshots, charts and documents and answer in text.

$ / 1K images

Models and APIs built for writing, completing and fixing code.

$ / 1M out tokens

Text-to-image and image-to-image models.

$ / image

Inpainting, background removal, upscaling, relighting.

$ / image

Consistent sprites, tilesets, icons and pixel art for games.

$ / asset

Text-to-video and image-to-video models.

$ / second

Talking-head avatars, dubbing and lip-sync.

$ / minute

Upscale to HD/4K, smooth motion, denoise and restore footage.

$ / video min

Cut people and objects out of video without a green screen.

$ / video min

Text- and image-to-3D meshes with textures.

$ / model

Turn video or a text prompt into skeletal 3D animation.

$ / anim min

Generate seamless PBR textures or re-skin 3D models from text and images.

$ / texture set

Natural speech, voice design and cloning.

$ / 1M chars

Transcription, diarization and real-time captions.

$ / audio hour

Songs, instrumentals and sound effects.

$ / minute

Convert one voice into another while keeping timing and emotion.

$ / audio min

Translate spoken video or audio into other languages, keeping the voices.

$ / source min

Generate footsteps, ambience, impacts and foley from a text prompt.

$ / audio min

Clean up speech, remove noise, split songs into vocals, drums and more.

$ / audio min

Dedicated GPU instances rented by the hour.

$ / GPU-hr

Peer-to-peer and interruptible GPUs at the lowest prices.

$ / GPU-hr

Run containers on GPUs that scale to zero.

$ / GPU-hr

Pay-per-token APIs for open-weight models.

$ / 1M out tokens

Managed training of models on your data.

$ / 1M train tokens

Non-GPU hardware for AI: TPUs, Trainium/Inferentia and dedicated CPU endpoints.

$ / accel-hr

Managed storage and similarity search for embeddings.

$ / mo · 1M vectors

Extract text, tables and fields from PDFs and scans.

$ / 1K pages

Search the live web from your app or agent.

$ / 1K queries

Hosted headless browsers and sandboxes for AI agents.

$ / browser-hr

Detect harmful content, PII and prompt injection.

$ / 1K requests

One API in front of many models, with fallbacks, caching and cost controls.

$ / mo · 1M requests

Tracing, prompt management and evaluation for LLM apps and agents.

$ / mo · 1M events

Managed runtimes and ready-made agents that do multi-step work for you.

$ / agent-hr

Personalised product, content and next-item recommendations as a service.

$ / 1K requests