How to Use Flux Kontext Pro on VdoBloom: Instruction Editing in 7 Credits

Flux Kontext Pro is Black Forest Labs’ instruction-based image model, and on VdoBloom it costs a flat 7 credits per image — about $0.35 at the Lite plan rate — whether you use it to generate an image from text or to edit one you upload. Its defining trick is preservation: you describe the single change you want, and it applies that change while leaving the rest of the frame — the face, the lighting, the background — intact. That is the behaviour most people actually want from an “AI editor” and the one most generators fail at.

What Flux Kontext Pro is genuinely best at

Kontext is the instruction-editing branch of the Flux family. Pro is its workhorse tier. Point it at an existing image with a plain-English instruction and it behaves like a retoucher rather than a re-generator. Three jobs where it consistently beats a general text-to-image model:

• Single-element swaps. Change a jacket from grey to burgundy, change the wording on a shop sign, remove a stray coffee cup from a table. The surrounding pixels stay put.
• Character consistency across a set. Instead of re-rolling a prompt and praying the same face comes back, you generate once and then edit forward — new pose, new outfit, new backdrop — from the same source frame.
• Ratio-flexible marketing assets. It outputs at 21:9 for ultrawide banners all the way down to 9:16 for vertical social, so one concept covers a whole placement list.

On VdoBloom it carries a 4.3 user rating and sits at the platform’s MODERATE content label — standard filtering, not the strict lockdown some models run. Full spec sheet lives on the Flux Kontext Pro model page.

Real specs and the exact credit cost

Credits on VdoBloom convert cleanly: the Lite plan is $15 for 300 credits, so one credit is $0.05. Kontext Pro is not resolution-tiered — there is one price, which makes budgeting simple.

Spec	Flux Kontext Pro
Developer	Black Forest Labs
Modes	Text-to-image and instruction editing
Credit cost per image	7 credits
Dollar cost per image	$0.35 at $0.05/credit
Reference images accepted	1
Aspect ratios	21:9, 16:9, 4:3, 1:1, 3:4, 9:16
Default ratio in the editor	16:9
Output formats	JPEG (default) or PNG
Prompt limit	3,000 characters
User rating	4.3
Content label	MODERATE

For scale: your 10 free signup credits buy one Kontext Pro image with 3 credits left over. A 300-credit Lite month is roughly 42 Kontext Pro renders. Credit packs start at $2.49 for 75 credits and never expire — the full ladder is on the pricing page.

How to prompt it well

Kontext Pro is an instruction model, not a description model. That one distinction fixes most bad results.

Write the change, not the scene. “A woman in a red coat standing in a rainy street” is a text-to-image prompt — hand it to Kontext with an uploaded photo and you invite it to rebuild the whole frame. “Change her coat to red” is a Kontext prompt, and it will touch the coat and nothing else.

Name what must survive. When an edit is large enough to tempt the model into redrawing, add a preservation clause: “Replace the background with a sunlit beach. Keep the subject’s face, pose, hair and clothing exactly as they are.” That sentence is doing real work.

One instruction per run. Three stacked changes in a single prompt is where consistency drifts. Run three 7-credit passes instead, chaining each output into the next upload. You pay 21 credits for a result you can actually steer — and you can stop at pass two if it already looks right.

Quote text literally. For signage and packaging, put the string in quotation marks: Change the sign to read “OPEN LATE”. Typography is where the Max tier pulls ahead, so if the rendered text has to survive close inspection, spend the extra 3 credits.

Choose PNG when the image will be edited again. JPEG is the default and fine for delivery; PNG avoids compounding compression artefacts across a chain of edits.

If your source photo contains a real, identifiable person, you need that person’s consent before you edit and publish their likeness — that applies to every model on the platform, not only this one.

A worked example: one product shot, four placements

Say you have a single 4:3 photo of a ceramic mug on a wooden table and you need a full ad set.

1. Open the image editor, pick Flux Kontext Pro, upload the mug photo.
2. Set the ratio to 16:9 and prompt: Extend the scene into a wide kitchen counter. Keep the mug, its glaze, and the table grain exactly as they are. 7 credits.
3. Take that output, re-upload it, set 9:16, and prompt: Recompose vertically with the mug in the lower third and clean empty space above for text. 7 credits.
4. Take the original again, set 1:1, and prompt: Change the mug glaze to matte black. Keep lighting, shadows and background unchanged. 7 credits.
5. Final pass on the black version: Add the words “NEW MATTE” on the wall behind, in clean sans-serif. 7 credits.

Four assets, 28 credits, $1.40 — and the mug is recognisably the same mug in all four, which is the entire point.

When to pick a different model instead

If you need…	Pick	Credits
Instruction editing, one source image	Flux Kontext Pro	7
The same thing at higher fidelity and stronger typography	Flux Kontext Max	10
To expand the canvas beyond the original frame	FLUX Outpainting	12
Mask-based object removal with a painted mask	FLUX Erase	10
To blend many reference images into one composite	A multi-image editor — Kontext Pro takes only 1 input	varies
The cheapest usable pass for rough exploration	FLUX.2 [dev]	4

The hard ceiling to remember is the single reference image. Anything that needs several sources fused — a person from one photo into a room from another — is the wrong job for this model, no matter how well you word it. Starting from nothing at all? Use image generation, where Kontext Pro also runs as a straight text-to-image model at the same 7 credits.

Frequently asked questions

How much does one Flux Kontext Pro image cost?

7 credits, flat, for both text-to-image and editing. At the Lite rate of $0.05 per credit that is $0.35 per image. There are no resolution surcharges.

Can Flux Kontext Pro take more than one reference image?

No. It accepts exactly one source image. For multi-reference composition you need a different editor on the platform.

What aspect ratios does it support?

Six: 21:9, 16:9, 4:3, 1:1, 3:4 and 9:16. The editor defaults to 16:9.

Should I use Kontext Pro or Kontext Max?

Pro for volume and iteration at 7 credits. Max for final-delivery hero assets and anything where rendered text has to hold up, at 10 credits. They share the same six ratios, so you can switch tiers without reframing.

Is my image watermarked?

Free accounts get watermarked output. Any paid plan downloads watermark-free with commercial rights.

How strict is the content filter?

Kontext Pro carries the MODERATE label — standard filtering rather than a strict lockdown. Explicit and illegal content stays blocked on every model.