One Account for LLM API plus Image & Video Generation

Billing units by model type

Text models: input, cache read and output are priced separately, per 1M tokens.

Image models: it depends on the model. GPT Image models such as gpt-image-2 are priced per 1M tokens; generated image tokens cost $5.01 on LanoX against a listed official price of $30.00. Other models are priced per image by resolution, such as qwen-image-3.0-pro below.

Video models: priced per output second, by resolution and input conditions. The prices below are for requests without video input.

Estimating a job

• Text: tokens ÷ 1,000,000 × price, for input and output separately.
• Image: price per image at your resolution × number of images. 10 qwen-image-3.0-pro images at 2K = $0.602.
• Video: price per output second at your resolution × seconds. 10 seconds of grok-imagine-video-1.5 at 720P = $0.77, against $1.40 at the listed official price.

One key, one base address

The LanoX docs cover text, image and video APIs under one base host, https://api.lanox.ai. Text models support OpenAI-compatible and Claude Messages formats; image and video APIs are grouped by provider and task type. Image models such as lanox-image-2 have their own guide.

Billing is in US dollars with no monthly fee, and every call shows up in the console usage record.

What to check first

• The billing unit of the exact model and mode you plan to use.
• Whether the price is listed. Some modes, such as grok-imagine-video at 1080P, show “Price pending”.
• The model IDs your key can use, from GET /v1/models.