gpt-image-2.5 batch pricing: $4 image input, $15 output per 1M

OpenAI's batch rate for gpt-image-2.5 is $4.00 image input and $15.00 output per 1M tokens. Sume's docs describe a per-image reserve, not a batch tier.

4 min readSume
All posts

On OpenAI's pricing page, the batch rate for gpt-image-2.5-sunburst and gpt-image-2.5-flare is $4.00 per 1M image input tokens and $15.00 per 1M output tokens, against $8.00 and $30.00 at standard. Sume does not list a batch tier in the sources cited here; its image jobs reserve a USD estimate on submit, and the catalog's estimate for Image 1.0 is $0.2225 per image.

Read 2026-10-01.

What are the posted token rates?

All figures are per 1M tokens, as listed for both gpt-image-2.5 variants on the page.

OpenAI gpt-image-2.5 rates per 1M tokens, read 2026-10-01.
ModalityTierInputCached inputOutput
ImageStandard$8.00$2.00$30.00
ImageBatch$4.00$1.00$15.00
TextStandard$5.00$1.25-
TextBatch$2.50$0.625-

How does Sume price an image?

The catalog basis says Image 1.0 is built on GPT Image 2.5 image tokens. Its estimate is a prompt-only request at high quality; the minimum is a low-quality 1408×480 request, and the maximum reserves max quality, auto size, 4 images and 16 references. The rate card describes routers as "Provider list rates × Sume margin."

When is the money reserved?

Per Generation admission, paid generation uses public Sume USD estimates, reserved when the request is accepted and captured on successful completion. To preview without submitting, use dry_run=true as described in MCP tools and gates.

Where can I read more on Sume's image price?

See gpt-image-2.5 API price for the per-request breakdown and flare vs sunburst for the two OpenAI variant ids.

Sources

Related posts

More in Pricing

All Pricing posts

Written by Sume