AI image cost per image: Gemini, Luma, Ideogram and GPT Image lists
Vendor lists per image today: Gemini Nano Banana 2 $0.067 (1K), Luma Uni-1.1 $0.0404 (2K), Ideogram 4.5 $0.03-$0.22, GPT Image 2.5 $0.09366 (xhigh, 1024).

Per image, the lowest list prices read today are Gemini 3.1 Flash Lite Image at about $0.0336 for a 1K image, Luma Uni-1.1 at $0.0404 for 2K, and Ideogram 4.5 at $0.03 on low quality; the high-quality end runs from $0.134 for Gemini 3 Pro Image to $0.22 for Ideogram 4.5 high. What Sume charges is shown per endpoint by the Image API, and this post states vendor lists only.
Sources: Gemini API pricing, Luma API pricing, fal's Ideogram 4.5 page as quoted in Sume's Image API docs, which also give the GPT Image 2.5 token rates.
The lists
Gemini and Luma rates are the vendors' own; Ideogram and GPT Image 2.5 are fal list rates quoted by Sume's docs.
| Model | Setting | Per image |
|---|---|---|
| Gemini 3.1 Flash Lite Image | 1K | ~$0.0336 |
| Gemini 3.1 Flash Image (Nano Banana 2) | 1K / 2K | ~$0.067 / ~$0.101 |
| Gemini 3 Pro Image (Nano Banana Pro) | 1K-2K / 4K | ~$0.134 / ~$0.24 |
| Luma Uni-1.1 / Max | 2K text to image | $0.0404 / $0.1000 |
| Ideogram 4.5 | low / medium / high | $0.03 / $0.06 / $0.22 |
| GPT Image 2.5 (fal token rates) | 1024x1024 xhigh / max | $0.09366 / $0.21072 |
How to read your Sume price
Sume's Image API docs state that endpoint pricing lines are the amount charged to your wallet with Sume's margin already applied, so cost_usd x n is what you pay. Read it from GET /v1/images/models/{id}/endpoints rather than computing from the lists above.
Budget 1,000 images by multiplying: at Gemini's ~$0.067 that is about $67 on the vendor list; at Luma's $0.0404, $40.40; at Ideogram medium, $60. These are vendor-list sums, not Sume quotes, and the GPT Image numbers exclude input tokens.
Sources
Related posts
More in Pricing
- Which AI image models cost under 4 cents on the Sume image API?
Six Sume image models list under $0.04 per image: Soul, Grok Imagine, Qwen Image, Imagen 4 Fast, Seedream 4.0 and Flux 2 Pro. What each takes, for drafts.
- AI lip sync API cost per second: H3 Max 480p, 768p, 1080p vs Fabric
On Sume, H3 Max lip-sync is $0.0625, $0.10 or $0.20 per audio second by resolution; VEED Fabric is $0.10 or $0.1875. A 14.8 s clip costs $0.94 to $3.00.
- AI music for 100 short videos: $12.50 flat on Sume Music
One hundred Sume Music generations cost $12.50 at the fixed $0.125 per accepted generation, whatever the prompt length. What that does and doesn't cover.
- AI video cost per deliverable: a rate sheet for finance approval
A cost-per-deliverable sheet built from Sume's published rates: 15-second avatar ad, captioned cut, 6-image set, voiced script, music bed, with a retake factor.
Written by Sume