API / v2

All models

Pick a model below. Each model exposes its own endpoints (text-to-image, image-to-image, etc.) at /api/v2/{model}{/operation}.

Showing 101–105 of 105

Speech to Speech preview

Speech to Speech

1 endpoint

Convert an existing speech recording into a different voice while preserving the original timing, prosody, and emotion. Works with built-in voices, premium presets, and custom cloned/designed voices.

Speech to Speech
/api/v2/speech-to-speech
Pricing 25 cr / call
No preview yet

Whisper (Transcription)

1 endpoint

Transcribe speech to timed segments. Used by footage indexing and captioning.

Transcribe
/api/v2/whisper
Sync
Pricing 1 cr / second
No preview yet

Gemini Flash (Video Understanding)

1 endpoint

Analyse a whole video natively (motion + audio) into a timed transcript and visual moments. Moderated — refuses explicit content (returns nsfw=true).

Analyze Video
/api/v2/gemini-flash
Sync
Pricing 1 cr / call
No preview yet

Qwen 3 VL

1 endpoint

Uncensored native video analysis (motion + audio). Use for unrestricted footage that moderated endpoints refuse.

Analyze Video
/api/v2/qwen3-vl
Sync
Pricing 1 cr / call
No preview yet

Text Embeddings

1 endpoint

Create embedding vectors for semantic search across transcript, captions and notes.

Embed Text
/api/v2/text-embedding-3-small
Sync
Pricing / call