Skip to content

    Capability

    Speech-to-text,in one request.

    Transcribe audio with open speech models on the OpenAI-compatible /v1/audio/transcriptions endpoint. Drop it into voice agents, meeting tools, and call analytics, billed per minute.

    Get an API keyAll models

    Models you can call

    Served on FlexAI today, on the OpenAI-compatible API.

    ModelContextPriceStatus
    Whisper Large V3 Turbo—$0.0006 / min Serving
    NVIDIA Parakeet TDT 0.6B v3—$0.0014 / min Serving

    Call it

    OpenAI-compatible /v1/audio/transcriptions. Same key as every other model.

    curl https://tokens.flex.ai/v1/audio/transcriptions \
  -H "Authorization: Bearer $FLEXAI_API_KEY" \
  -F model="whisper-large-v3-turbo" \
  -F file=@audio.mp3

    In the pipeline

    Use cases that put this capability to work.

    Support agents

    Agents that triage, answer, and resolve customer conversations.

    See the pipeline

    Start serverless and pay per use. When volume proves out, move to a dedicated endpoint on the same key.

    Every modality, one API key

    Text, vision, image, audio, and embeddings behind one OpenAI-compatible key.

    Get an API keyTalk to us

    $10/month in free credits for your first 3 months

    See how much you could save