echotokens docs

Build with the gateway.

Drop-in OpenAI SDK shape for image, video, and voice. Flat USD pass-through pricing — every response includes the exact cents we charged your wallet.

Quickstart

Drop in the OpenAI SDK base URL and ship in five minutes.

Images

Generate and edit images. OpenAI-compatible request shape, signed asset URLs.

Video

Text-to-video and image-to-video with Veo, Kling, and Runway. Async polling.

Text-to-speech

Deepgram Aura and ElevenLabs. Returns binary audio you can stream.

Transcription

Deepgram Nova-3 / Nova-2. Multipart upload up to 25MB.

Errors

Standard error envelope, status codes, and how to surface them.

What is echotokens?

echotokens is a media-model gateway with flat USD, pass-through pricing. One endpoint for image, video, and voice — drop-in OpenAI SDK compatible. Every response includes a cost_usd_cents field that's exactly what your request cost us; we take no markup.

python · first image
from openai import OpenAI

client = OpenAI(
    base_url="https://api.echotokens.me/v1",
    api_key="sk-echo-...",
)

res = client.images.generate(
    model="nano-banana-2",
    prompt="a serene mountain lake at dawn",
    n=1,
)
print(res.data[0].url)
# Every response includes `cost_usd_cents` — pass-through.

Three things to know before you ship

  1. Base URL. https://api.echotokens.me/v1 — point your OpenAI SDK at it and the rest of your code keeps working.
  2. Auth. Every request needs a Bearer token. Generate one on the keys page — it looks like sk-echo-.... Treat it like a password.
  3. Asset URLs. Image and video results come back as signed URLs on our host (/v1/assets/<token>). They're valid for 24 hours; download or re-mint within that window.

Surface map

Pick the surface you need — each guide is short and tied to a working studio page you can try without writing code:

  • Quickstart — first generation in three lines (cURL, Python, Node).
  • Images — text-to-image and image edits.
  • Video — text-to-video and image-to-video with polling.
  • Text-to-speech — Deepgram Aura + ElevenLabs voices.
  • Transcription — Deepgram Nova-3 / Nova-2.
  • Errors — the standard error envelope and status codes.
model ids match the live catalog

Every model id and parameter listed here is also exposed live on the pricing page. If something on this page diverges from what the catalog reports, the catalog is the source of truth — file an issue so we can fix the doc.

Pricing

Top up $50 in crypto, receive $100 of credit. We bought the volume; you keep the savings. There's no monthly minimum and no platform fee — the only number that ever decrements your wallet is the per-request cost.

See the pricing page for live per-model rates, or visit billing/topup to add credit.