FLUX.1 Kontext Image Editor
FLUX.1 Kontext is Black Forest Labs’ previous-generation image-editing family. BFL recommends FLUX.2 for new projects, but Kontext remains available through its API and this site’s configured fal.ai route for workflows that start with an uploaded image.
提示词开头: Preserve the product shape, exact color, label placement, and camera angle. Replace only the background with charcoal stone, add a soft banana-amber rim light and realistic contact shadow, remove clutter, and create a premium 16:9 ecommerce hero with copy space on the right.
Flux 1 Kontext image editor
多模型生成，无需 API 密钥，可领取免费额度。
Last verified: 2026-08-22
Generation status and workflow guidance were checked against Black Forest Labs documentation and the fal.ai model route configured in this product.
生成前先看已核验事实
以下信息来自当前产品配置与所链接的一手文档。实际输出质量仍会受提示词与输入图像影响。
- BFL generation
- Previous generation
- Banananano route
- fal.ai · flux-pro/kontext
- Core workflow
- Prompt + input image
- BFL use cases
- Object, text, style edits
Black Forest Labs describes FLUX.1 Kontext as previous-generation and recommends FLUX.2 for new projects.
The current product configuration uses the fal-ai/flux-pro/kontext endpoint—not a FLUX.2 endpoint relabeled as Kontext.
Kontext editing begins with an image and an instruction. Upload is required for the editing workflow.
BFL documents targeted object, text, character, and style changes while using visual context from the source image.
Use Kontext for an explicit edit
Start with a source image you own or license. Write separate preserve and change instructions so unintended edits are easy to spot.
- Preserve subject identity, product geometry, label position, or pose
- Change background, lighting, style, object, text area, or crop
- State the destination: listing, ad, hero image, portrait, or social post
Do not confuse legacy availability with a recommendation
This model can still be useful for an established workflow, but new projects should also test the current FLUX.2 family or another available editor.
- Keep Kontext only when it performs better on your real edit set
- Compare with identical images, prompt, crop, and review criteria
- Check the live generator for provider availability and credit estimate
Review every supposedly preserved detail
“Preserve” is an instruction, not a guarantee. Identity, product details, small text, proportions, and shadows can still drift.
- Overlay the source and output to inspect geometry and identity changes
- Verify logos, packaging, text, hands, reflections, and added objects at full size
- Do not upload confidential, biometric, or unlicensed material
Measure edit fidelity instead of visual novelty
A good edit changes the requested region while protecting everything else. Score both sides of that requirement.
- 1Choose a rights-cleared source and list the elements that must not change.
- 2Write one edit instruction and use the same source, ratio, and settings for every run.
- 3Generate several outputs and keep failed edits in the review set.
- 4Score requested change, unintended change, identity or product fidelity, text, and publishability.
- 5Repeat with a current alternative before deciding to keep a legacy workflow.
This page does not publish a controlled Kontext benchmark or claim it outperforms FLUX.2. The visible model artwork is illustrative.
一手资料
- FLUX.1 Kontext overview — Black Forest Labs
- FLUX.1 Kontext image editing guide — Black Forest Labs
- What is FLUX.1 Kontext? — Black Forest Labs Help Center
常见问题
Is Flux 1 Kontext for text-to-image or editing?
This page and product route focus on image-to-image editing: upload a source image, then describe what to preserve and what to change.
Can it improve product photos?
It can be tested for background, lighting, and scene edits, but exact product fidelity is not guaranteed. Compare the output with the real product before using it commercially.
Is FLUX.1 Kontext the newest FLUX editor?
No. Black Forest Labs calls FLUX.1 Kontext previous-generation and recommends FLUX.2 for new projects. Kontext remains available for existing API and playground workflows.
开始使用 Banananano AI 生成图像
使用每日免费额度测试模型、保存提示词片段，并建立可复用的创作工作流。