manyframeFLUX 1.1 Pro.
Build a text-led product scene through a declared FLUX prediction route, with explicit composition and prompt-processing choices.
Available models and settings are shown in your studio. Review the quote before generating.
The route.
The right inputs.
Start from the creative job, then choose the model and settings that fit it.
Model ID black-forest-labs/flux-1.1-pro
- Request path
- Replicate model predictions with a required non-empty prompt
- Composition
- Standard size maps to a ratio or custom width and height
- Prompt upsampling
- quality set to hd or high enables prompt_upsampling
- Output fields
- output_format and positive n map into prediction inputs
Capabilities describe the Manyframe connector. Your studio shows the currently enabled route, supported settings and maximum charge.
Know what the quality setting changes.
For this connector, hd or high quality enables prompt upsampling. That is a prompt-processing choice rather than a promise of higher pixel dimensions. If you are comparing a tightly written brief, record whether upsampling was enabled so you can distinguish the submitted direction from the route’s handling of it.
Describe a product scene with a clear camera angle and separation from the background. The adapter can forward an uploaded edit image as image_prompt, but the configured model must accept that field. Confirm compatibility before using it to represent an existing product, and inspect the final details before approval.
Good questions. Clear answers.
Still curious? Browse all FAQs
Can I send an empty prompt and rely only on the image?
No. The Replicate image converter requires a non-empty prompt, including for the compatible edit forwarding path.
Does the Pro label guarantee a specific resolution or quality?
No such guarantee is established here. Confirm the provider’s supported dimensions, inputs and quote for the selected prediction route.
Are extra prediction inputs accepted?
The connector merges additional input fields into the prediction payload. This does not prove the model accepts every field, so verify the configured route’s schema before sending them.
Give the prediction a precise brief.
Select the FLUX route and check its composition and prompt settings.