Skip to content

Thousands of files, one API key

Images, video and voice in bulk from your code: one HTTP API and a hosted MCP server over every model in the catalog, a price per call, and a ref for every file so your pipeline can find it again.

From

$90.00

per piece at current prices

1,000 images on Seedream 4.0

What you can make

  • Thousands of images for datasets, catalogues or ad tests
  • Hundreds of short clips for feeds and templates
  • Hours of voiceover for courses, audiobooks or explainers
  • One prompt across several models to compare
  • Scheduled jobs that run without anyone watching

The pipeline

  1. 1

    Keys

    Create one API key per project or per job. Each key has its own spend, an alert and a one-click revoke.
  2. 2

    Requests

    POST /api/images/generate with a provider, /api/scenes/generate with an engine, /api/voiceover/generate for voice.
  3. 3

    Poll and fetch

    Every result carries a ref. GET /api/generations/{ref} returns its status and a fresh link to the file.
  4. 4

    Reconcile

    Each job is itemised with its model and its cost; a failed render goes back to the balance automatically.

Cost per piece

Each line is a real row of the rate table at the public prepaid price, multiplied out. A range means the price depends on something only the run decides, such as the length of the text. Each job shows what it actually cost.

1,000 images on Seedream 4.0

  • 1,000 × imagesSeedream 4.0 · $0.090 per image$90.00
Total · at current prices$90.00

100 ten second clips on Hailuo 02

1,000 seconds of video.

  • 1,000 × seconds of videoHailuo 02 · $0.090 per second$90.00
Total · at current prices$90.00

One hour of ElevenLabs voiceover

About 54,000 characters, 120 blocks.

  • 120 × voiceover blocksElevenLabs · $0.11 per 450 characters$12.96
Total · at current prices$12.96

Every rate, machine-readable: GET /api/pricing

Run it from Claude Code

Connect the hosted MCP server once, then describe the batch in plain words. Claude plans it, calls the models, waits for each render and hands you the files with what they cost.

claude mcp add --transport http aitachyon \
  https://aitachyon.com/api/mcp \
  --header "Authorization: Bearer ait_..."

“Read prompts.txt and generate one Seedream image per line at 1024 by 1024. Tell me the total before you start, and write each ref next to its prompt in results.csv.”

Or from your code

Plain HTTP with a bearer key. Video returns a job you poll; every file carries a ref you can fetch later with GET /api/generations/{ref}.

while IFS= read -r prompt; do
  curl -s https://aitachyon.com/api/images/generate     -H "Authorization: Bearer $AITACHYON_KEY"     -H "Content-Type: application/json"     -d "$(jq -n --arg p "$prompt"       '{projectId: "<project id>", prompt: $p, provider: "seedream"}')"
done < prompts.txt

curl https://aitachyon.com/api/generations/img_<uuid>   -H "Authorization: Bearer $AITACHYON_KEY"

Built for volume

  • Pay per call from a prepaid dollar balance that never expires
  • A failed render is refunded to your balance automatically
  • Per-key spend, alerts and one-click revoke, so a runaway script stays contained
  • Every file has a ref you can fetch with GET /api/generations/{ref}
  • The web studio, the HTTP API and the MCP server share one account and one balance

Platform rules you need to respect

Generating is the easy part. What you publish is your responsibility, and each platform has rules for AI content. Never present an AI person as a real customer, a real review or a real event.

  • YouTubeMonetized channels are reviewed for inauthentic content: mass-produced or repetitive videos without real value added. Realistic altered or synthetic content has to be disclosed when you upload.Channel monetization policiesDisclosing altered or synthetic content
  • TikTokRealistic AI-generated content has to carry a label, set with the AI-generated content toggle when you post. TikTok may also label it automatically.AI-generated content on TikTok
  • Meta (Facebook, Instagram)Meta labels content it detects as AI-made and asks you to disclose photorealistic video and realistic audio created or altered with AI. Ads about social issues, elections or politics have their own disclosure rules.Meta's approach to labeling AI content

Policies change: read the current version before you publish. We make the files; we do not promise monetization, reach or views.

Questions

At current prices, 1,000 Seedream images come to about $90.00. Every call is billed at the price listed for its model, and each job shows what it actually cost.

The amount it took returns to your balance automatically. Your script reads the status with GET /api/generations/{ref} and can retry.

Give each script its own key. You see what it spends, get an alert when it runs hot and can revoke it in one click.

Yes. The hosted MCP server exposes the same tools to Claude Code: describe the batch in plain words and it calls the models, waits for the files and reports the cost.

One account for all of it

Every model on one prepaid dollar balance, priced per call. Start in the studio, then move the same pipeline to Claude Code or your backend.

Start generating