MODELS

Available modalities

We expose standard inference endpoints across eight modalities. Usage allotments are scoped to each project; rate limits apply at the domain scope.

AllLiveSoonTBC

All modalities remain reachable with JavaScript disabled.

Chat · Summarize · Code

Standard chat completions endpoint — for chat, summarization, code generation, and tool use.

Learn more →

Search · Similarity · RAG

Turn text into vectors for semantic search, similarity, and RAG pipelines.

Learn more →

Reorder · Relevance scoring

Reorder search results by relevance and lift top-K precision.

Learn more →

Transcripts · Captions · Notes

Convert speech to text — for transcripts, captions, call analytics, and voice input.

Learn more →

Narration · Readout · Voice agents

Render text as natural-sounding audio — for narration, readouts, and voice agents.

Learn more →

Generate · Edit · Restyle

Create new images from text or images, and edit or restyle existing ones.

Learn more →

Video gen · Motion synthesis

Generate video clips from text or images — from short social cuts to long-form output.

Learn more →

Image understanding · OCR · Docs

Multimodal chat that accepts images, plus document OCR and table extraction.

Learn more →