OpenAI · fal.ai · Text to image
GPT Image 2
GPT Image 2 (OpenAI), text to image, served by fal.ai. The OpenAPI schema for the openai/gpt-image-2 queue. Price: Billed on image tokens; fal's canonical table gives per image (low / medium / high): 1024x768 $0.005 / $0.037 / $0.145, 1024x1024 $0.006 / $0.053 / $0.211, 1024x1536 $0.005 / $0.042 / $0.165, 1920x1080 $0.005 / $0.040 / $0.158, 2560x1440 $0.007 / $0.056 / $0.222, 3840x2160 $0.012 / $0.101 / $0.401. Presets other than square_hd render at 0.66 to 0.79 megapixel (measured 2026-09-04: square 816x816, 4:3 1024x768, 16:9 1088x608) and are priced at the 1024x768 tier; explicit sizes are the canonical ones in either orientation. Text tokens $5 per 1M input: $0.002 allowance per call. 'auto' size and quality are not offered: not priceable in advance.
Usage based fal.aiImagefal/gpt-image-2Listed
Activation pending; your agent will see it as unavailable until then.
Use this tool with your agent
Use my connected Agentik MCP server. I want to use the tool fal/gpt-image-2 (GPT Image 2 (OpenAI), text to image). Call inspect_tool with {"tool_id": "fal/gpt-image-2"} first and tell me the price before anything runs. Ask me for the inputs it needs; never invent values, IDs or file URLs. When I confirm, call run_tool with tool_id "fal/gpt-image-2" and my inputs, follow it with get_run if it is still running, and give me the result and the price Agentik reports. If the tool is unavailable, say so and use discover_tools to propose an alternative.Show the prompt
Paste it into Claude, ChatGPT or Codex: your agent takes it from there, and shows the price before it runs.
Parameters
| Name | Type | Required | Description |
|---|---|---|---|
num_images | integer | No | Number of images to generate |
prompt | string | Yes | The prompt for image generation |
output_format | string | No | Output format for the images Options: jpeg, png, webp. |
background | string | No | Background for the generated image Options: auto, transparent, opaque. |
quality | string | No | Quality for the generated image. Use 'auto' to let the model pick the best quality for the prompt. Options: low, medium, high. |
image_size | string | No | The size of the generated image. Supports preset names, explicit {width, height}, or 'auto' to let the model pick the best size. Concrete sizes must have both dimensions as multiples of 16, max edge 3840px, aspect ratio <= 3:1, total pixels between 655,360 and 8,294,400. Options: square, portrait_4_3, portrait_16_9, landscape_4_3, landscape_16_9, 1024x768, 768x1024, square_hd, 1024x1024, 1024x1536, 1536x1024, 1920x1080. |
How your agent calls it
First inspect_tool for the current schema and price, then run_tool with the published example after confirming your inputs:
{
"tool_id": "fal/gpt-image-2",
"input": {
"prompt": "a small red cube on a white table",
"quality": "low",
"image_size": "landscape_4_3",
"num_images": 1
}
}Price and billing
- Price
- Usage based
- Model
- Per unit
- Detail
- The provider’s price, with no markup. Billed for measured usage.
- The provider’s price, 0% markup.
- One prepaid balance for every provider; your agent sees the price before it runs.
- Pay for measured usage; unused reservations are released.