flux-2/max/text-to-image
Create photoreal visuals with multi-reference, color, and typography precision.
Black Forest Labs' FLUX 3 Image turns a written prompt into a detailed, well-composed picture, with native output at 1K, 2K, and 4K and cleaner text rendering.
The same image family includes FLUX 3 Image Edit, which makes precise local edits and can combine up to 10 reference images. This RunComfy model calls text-to-image only.
Trading manual layout and retouching for a prompt, FLUX 3 Image helps designers, marketers, and product teams ship stills faster.
For developers, FLUX 3 Image on RunComfy can be used both in the browser and via an HTTP API, so you don't need to host or scale the model yourself.
Ideal for: Campaign-Ready Product Imagery | Layout-Accurate Posters and Ads | High-Resolution Still Frames
FLUX 3 Image is Black Forest Labs' text-to-image model. It generates detailed, well-composed pictures from a prompt, with native output up to 4K and cleaner text rendering. On RunComfy you can iterate in the model UI and reuse the same settings over HTTP API.
FLUX 3 Image Edit is the companion model in the same family. It makes precise local edits (add, replace, remove, or restyle one element) and can combine up to 10 reference images. This page does not call that endpoint. Generation here is text-to-image only.
Output format: Image / jpeg or png / resolution tiers 1k, 2k, and 4k
| Parameter | Required | Type | Default | Range / Options | Description |
|---|---|---|---|---|---|
| prompt* | Yes (*) | string | — | — | Describe the subject, composition, lighting, style, and any text to include. |
| aspect_ratio | No | string | 1:1 | 21:9, 2:1, 16:9, 3:2, 7:5, 4:3, 5:4, 1:1, 4:5, 3:4, 5:7, 2:3, 9:16, 1:2 | The aspect ratio of the generated image. |
| resolution | No | string | 1k | 1k, 2k, 4k | Output resolution tier. Higher tiers cost more; 4K can take several minutes. |
| output_format | No | string | jpeg | jpeg, png | The format of the generated image. |
| enable_prompt_expansion | No | boolean | false | true, false | Expand the prompt while preserving its intent and the roles of reference images. |
Image edit is part of the same FLUX 3 image release. It is described here so the family is clear. This RunComfy tool does not send image_urls and does not call that endpoint.
RunComfy charges $0.055 per 1K image, $0.12 per 2K image, and $0.68 per 4K image.
1) Open FLUX 3 Image on RunComfy and write a prompt for subject, setting, light, and any on-image text.
2) Set Aspect Ratio. The default is 1:1; other options include 16:9, 9:16, 21:9, and 1:2.
3) Choose Resolution: 1k, 2k, or 4k. Start at 1k while iterating, then step up for delivery.
4) Pick Output Format: jpeg by default, or png when you want a lossless file.
5) Turn on Enable Prompt Expansion only when a short prompt needs a fuller description.
6) Generate, review the still, then refine the prompt before raising resolution.
7) For API use on RunComfy, send the same parameters. This route is text-to-image, not image edit.
In short, FLUX 3 Image on RunComfy turns prompts into stills up to 4K. FLUX 3 Image Edit is the family's editing model; this tool calls text-to-image only.
Create photoreal visuals with multi-reference, color, and typography precision.
Create multilingual, high-fidelity visuals with precise text-driven generation and seamless edit control.
Refine texture, geometry, and lighting with chrono-edit upscaler for realistic image upscaling.
Transforms images into editable RGBA layers for precise object isolation and seamless design control.
Fast, high-quality text-to-image generation with Nano Banana 2, with aspect ratio and resolution controls.
Generate branded visuals with accurate in-image text and logos.
Yes. FLUX 3 Image generates a still from a prompt. FLUX 3 Image Edit is a separate model in the same family for local edits and reference images; it is not called from this tool.
FLUX 3 Image turns a text prompt into a detailed, well-composed picture, including native 2K and 4K tiers and cleaner on-image text. It fits posters, product stills, ads, and other layout-sensitive graphics.
FLUX 3 Image Edit makes precise local changes and can combine up to 10 reference images. Its required inputs are a prompt and image URLs. That endpoint is not wired on this RunComfy model, which stays text-to-image.
Yes. Put the exact words in quotes and describe their style and placement. Legibility still depends on the prompt, the resolution tier, and the layout you ask for.
prompt is required. Optional fields are: aspect_ratio (default 1:1; 21:9, 2:1, 16:9, 3:2, 7:5, 4:3, 5:4, 1:1, 4:5, 3:4, 5:7, 2:3, 9:16, 1:2), resolution (1k, 2k, 4k; default 1k), output_format (jpeg or png, default jpeg), and enable_prompt_expansion (default false). There is no 512sq or 768sq tier, no safety tolerance, no sync mode, and no seed.
Prototype in the RunComfy model UI, then call the same template through the RunComfy API with the same input fields. The result is an image URL.
RunComfy charges $0.055 per 1K image, $0.12 per 2K image, and $0.68 per 4K image.
RunComfy is the premier ComfyUI platform, offering ComfyUI online environment and services, along with ComfyUI workflows featuring stunning visuals. RunComfy also provides AI Models, enabling artists to harness the latest AI tools to create incredible art.
