Models / Media / Text to image
Text to image API. 26 models from Google, OpenAI, ByteDance, Black Forest Labs, Kling and Alibaba on one key, priced per image.
Prepaid credit, no subscription. A generation that fails is not charged. The list below comes from the live catalog, so it changes as models are added.
Models · lowest price per image, USD, next to the developer's own price where on record
| Model | Developer | UplinkSwitch | Developer price |
|---|---|---|---|
Nano Banana 2nano-banana-2 | from $0.0402 / image | $0.067 | |
Nano Banana Pronano-banana-pro | from $0.0804 / image | $0.134 | |
Nano Banananano-banana | from $0.0313 / image | — | |
Nano Banana 2 Litenano-banana-2-lite | from $0.0313 / image | $0.0336 | |
Nano Banana 2.1nano-banana-2.1 | from $0.025 / image | $0.0336 | |
GPT Image 2gpt-image-2 | OpenAI | from $0.024 / image | — |
GPT Image 2.5 Flaregpt-image-2.5-flare | OpenAI | from $0.0197 / image | — |
GPT Image 2.5 Sunburstgpt-image-2.5-sunburst | OpenAI | from $0.0197 / image | — |
GPT Image 1.5gpt-image-1.5 | OpenAI | from $0.0125 / image | — |
GPT-4o Imagegpt-4o-image | OpenAI | from $0.06 / image | $0.10 |
Seedream 5.0 Proseedream-5.0-pro | ByteDance | from $0.0938 / image | $0.0938 |
Seedream 5.0 Liteseedream-5.0-lite | ByteDance | from $0.0313 / image | — |
Seedream 5.0 Flashseedream-5.0-flash | ByteDance | from $0.06 / image | — |
Seedream 4.5seedream-4.5 | ByteDance | from $0.0313 / image | — |
Seedream 4seedream-4 | ByteDance | from $0.075 / image | — |
FLUX.2 Proflux-2-pro | Black Forest Labs | from $0.0563 / image | — |
FLUX.2 Flexflux-2-flex | Black Forest Labs | from $0.1688 / image | — |
FLUX.1 Kontext Proflux-kontext-pro | Black Forest Labs | from $0.12 / image | — |
FLUX.1 Kontext Maxflux-kontext-max | Black Forest Labs | from $0.24 / image | — |
Kling O3 Imagekling-o3-image | Kling | from $0.0219 / image | — |
Qwen Image 3qwen-image-3 | Alibaba | from $0.03 / image | — |
Qwen Image 3 Proqwen-image-3-pro | Alibaba | from $0.04 / image | $0.04 |
Qwen Image 2.1qwen-image-2.1 | Alibaba | from $0.0375 / image | $0.04 |
Wan 2.7 Imagewan-2.7-image | Alibaba | from $0.0262 / image | — |
Wan 2.7 Image Prowan-2.7-image-pro | Alibaba | from $0.0656 / image | — |
Z-Imagez-image | Alibaba | from $0.03 / image | — |
Each model's page lists the price for every setting. Live from GET https://api.uplinkswitch.com/v1/catalog/media.
How to call it · POST /v1/images/generations
Image models answer POST /v1/images/generations the way OpenAI does, so the OpenAI SDK works with a new base URL. Model-specific settings go in input. Every model here also takes POST /v1/generations. The media guide has the full flow. This request uses Nano Banana 2; swap the model id for any row above.
curl
curl https://api.uplinkswitch.com/v1/images/generations \
-H "Authorization: Bearer $UPLINK_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "nano-banana-2",
"prompt": "A red paper boat on a calm blue pond, morning light"
}'Questions
- What does text to image cost?
- From $0.0125 per image (GPT Image 1.5). Each model's page lists its price for every setting. The price is fixed when the request is accepted, and there is no subscription.
- How long does it take?
- Usually 10 to 40 seconds. A call to POST /v1/images/generations waits for the image; POST /v1/generations answers at once with a generation id to poll.
- What if a generation fails, and how long are files kept?
- A generation that fails is refunded in full, automatically. Files stay downloadable for 30 days at a URL on api.uplinkswitch.com.