Black Forest Labs/flux-image-upscaler/Released Aug 2026

FLUX Image Upscaler

Detail-preserving image upscaling up to 4x

Commercial use

Image

About FLUX Image Upscaler

FLUX Image Upscaler enlarges a still up to 4x while adding detail, holding natural color and keeping artifacts down. It suits photographs, product shots, and generated images that need to print or display larger than they were made.

Creativity controls how far the model may depart from the source: low values restore detail conservatively, higher values invent texture where the original has none.

See what this model can do

A handful of real prompts and their outputs.

Portrait Detail

Sailor's Hands

"Recovers skin creases, knuckle wrinkles and individual rope fibers from a thumbnail-sized source."

Landscape

Mossy Stone Bridge

"Rebuilds stone joints, moss texture, ferns and moving water while keeping the original colors."

Side-by-side crop of a vintage film camera lens and worn desk, soft before and detailed after 4x upscaling.

Product

Vintage Camera

"Restores knurling on the lens barrel, leatherette grain and wood scratches without inventing new objects."

Side-by-side crop of a tabby cat's eye and nose, blurry before and sharp after 4x upscaling.

Pets

Tabby Close-Up

"Brings back individual fur strands, whiskers and iris detail at low creativity."

Side-by-side crop of ivy on a red brick wall, soft before and crisp after 4x upscaling.

Texture

Ivy on Brick

"Separates every ivy leaf and its veins and restores crumbling mortar lines between the bricks."

Wildlife

Great Horned Owl

"Rebuilds fine barred feather patterns and crisp iris rims around the owl's yellow eyes."

Side-by-side crop of an embroidered mountain patch on denim, soft before and thread-sharp after 4x upscaling.

Macro Detail

Embroidered Patch

"Resolves individual satin-stitch threads, the merrowed border and the denim weave."

Side-by-side crop of alpine chalets on a hillside, blurry before and detailed after 4x upscaling.

Landscape

Alpine Village

"Sharpens chalet balconies, roof shingles, chimney smoke and fence lines from tiny source pixels."

Side-by-side crop of wet strawberries, soft before and glossy and sharp after 4x upscaling.

Food

Fresh Strawberries

"Brings back seed texture, specular highlights and water droplets on the fruit."

Ready to build with FLUX Image Upscaler?

Try FLUX Image Upscaler in the Workbench to prompt it, compare outputs, and iterate on prompts without writing any code. When you're ready to ship, call the same model from our API and build your own apps on top of it.

curl -sSf -X POST https://hub.oxen.ai/api/ai/images/edit \
    -H "Content-Type: application/json" \
    -H "Authorization: Bearer $OXEN_API_KEY" \
    -d '{
  "model": "flux-image-upscaler",
  "input_image": "https://hub.oxen.ai/api/repos/ox/Oxen-AI-Assets/file/main/images/ox_zoom_out_1926_1076.png",
  "upscale_factor": 2,
  "creativity": 0.3,
  "num_inference_steps": 20
}'

Request parameters

Every field FLUX Image Upscaler accepts in the request body. See the API reference for response formats.

POSThttps://hub.oxen.ai/api/ai/images/edit

Field	Type	Required	Default	Description
`input_image`	`string` uri	Yes	—	Image to upscale. Must be jpg, jpeg, png, webp, gif, or avif.
`upscale_factor`	`number`	No	`2`	How much to enlarge the image. Cost follows the output area, so 4x costs roughly four times 2x. Range: 1 – 4
`creativity`	`number`	No	`0.3`	How far the model may depart from the source. Low values restore detail conservatively, high values invent new texture. Range: 0 – 1
`num_inference_steps`	`integer`	No	`20`	Number of denoising steps. More steps refine detail and take longer. Range: 4 – 50
`seed`	`integer` nullable	No	—	Random seed. Set for reproducible output.

Model JSON Image editing reference Async queue Models endpoint