Image Generation

The image models currently available in Langdock include Flux1.1 Pro Ultra and Flux.1 Kontext from our partner Black Forest Labs. Additionally, you can access Imagen 4, Imagen 4 Fast, and Gemini 2.5 Flash Image (Nano Banana) from Google, as well as DALL·E 3 and GPT Image 1 from OpenAI.

You can select any language model for image generation. Each model sends prompts to the underlying image generation model differently, so feel free to try different models and see how the generated images differ.

For guidance on writing good prompts, see the Prompt Engineering Guide: https://docs.langdock.com/resources/prompt-elements

1

How image generation works

You can use image generation in Langdock via the “Image” button in the chat field. This will use the default image model. You can also select a different model using the selector in the button.

2

Background prompt generation

The chat model will then choose the image generation tool and write a prompt to the image model in the background.

3

Image creation and return

The image model generates the image based on the prompt and returns it to the main model and you as the user.

Text in generated images may include mistakes or non-existent letters. This occurs because image models are trained on real images that include text; they mimic letter-like shapes but cannot reliably render correct sentences. This is a known limitation being actively improved by model developers.