Image API usage shows 0 tokens: where the cost is on Sume

In Sume's POST /v1/images response, usage.prompt_tokens, completion_tokens and total_tokens are always 0 in v1, and usage.cost is the USD amount billed to your wallet. Image models are metered per image, and per-token accounting is not plumbed through yet.

From Sume's Image API docs, read 2026-09-30.

What does the usage block look like?

The documented response for a Seedream 4.5 example:

{
  "created": 1748372400,
  "model": "bytedance-seed/seedream-4.5",
  "data": [{ "url": "https://media.sume.com/img/01J.../0.png", "media_type": "image/png" }],
  "usage": {
    "prompt_tokens": 0,
    "completion_tokens": 0,
    "total_tokens": 0,
    "cost": 0.04
  }
}

Which field do I record?

Record usage.cost. The model field echoes the id you requested, so sume/auto stays sume/auto. Endpoint pricing lines are the amount charged with Sume's margin already applied, so cost_usd x n is what you pay for n images.

Why does GPT Image 2.5 still mention tokens?

Sume's docs describe ChatGPT Image 2.5 pricing from Fal token rates and estimated token counts, but the response still reports per-image cost. The docs say the response token counts are always 0 in v1, so use usage.cost.

What about failed generations?

Billing is all-or-nothing: completed generations are billed in full, failed or cancelled ones are not billed, and failed requests return 502. A 202 job returns the result later from GET /v1/jobs/{id}/result, in the standard job shape rather than this body.

How do I check this myself?

If your dashboards multiply tokens by a rate, they will show zero for images. Sum usage.cost per request instead. The linked docs pages and the catalog endpoint show the current values, and this post reflects them as of 2026-09-30.