bfl/flux-2/klein/realtime
Flux 2 Klein Realtime is BFL's intelligent image editing model. Transform, retouch, and reimagine existing images using text prompts - from background replacement to artistic style conversion.
PNG, JPEG, WebP, or GIF · 20 MiB maximum
Example output — click Run to generate your own
Send your first request
OpenAI-compatible endpoint with unified authentication and usage tracking.
https://api.sandbase.ai/v1/runbfl/flux-2/klein/realtimeInput Schema
8 parameters · 2 required · 6 optional
| Parameter | Type | Required | Description |
|---|---|---|---|
image | string | Required | Base64-encoded image data URI for editing. CDN URLs are not supported for realtime. For optimal performance, use 704x704 JPEG images with 50% quality. Other sizes will be resized automatically. |
prompt | string | Required | The prompt to guide image editing. · Default: "Turn this into \"Living oil painting, melting gold and sapphire\"" |
seed | integer | Optional | Random seed for reproducibility. · Default: 35 |
schedule_mu | number | Optional | Schedule mu for time shift. 2.3=default, lower=more even denoising, 0.3=nearly linear. · Min: 0.3 · Max: 2.5 · Default: 2.3 |
aspect_ratio | string | Optional | The aspect ratio of the generated image. · Options: 21:9, 16:9, 3:2, 4:3, 5:4, 1:1, 4:5, 3:4, 2:3, 9:16 21:916:93:24:35:41:14:53:42:39:16 |
num_inference_steps | integer | Optional | Min: 1 · Max: 8 · Default: 3 |
enable_interpolation | boolean | Optional | Enable RIFE frame interpolation between consecutive frames (doubles output frames). · Default: false |
output_feedback_strength | number | Optional | Output feedback loop. 1.0 = pure noise (no feedback), 0.9 = 90% noise + 10% previous output latent. · Min: 0 · Max: 1 · Default: 1 |
Output Schema
| Field | Type | Description |
|---|---|---|
id | string | Unique identifier for the generation task |
status | string | Task status: pending, running, completed, failed, timeout |
model | string | Model used for the generation |
outputs | array | Array of output items |
outputs[].url | string | URL of the generated artifact |
outputs[].content_type | string | MIME type (e.g. image/png, video/mp4) |
error | object | null | Error details if failed, null on success |
error.type | string | Machine-readable error type code |
error.message | string | Human-readable error description |
Async Workflow
This model uses asynchronous execution. Submit a request and poll for the result.
- Submit — POST to /v1/run, receive an
id - Poll — GET /v1/run/{id} until status is
completed,failed, ortimeout - Retrieve — Read
outputsfrom the completed response
Code Examples
Ready-to-run snippets
# Step 1: Submit
curl -X POST https://api.sandbase.ai/v1/run \
-H "Content-Type: application/json" \
-H "Authorization: Bearer YOUR_API_KEY" \
-d '{
"model": "bfl/flux-2/klein/realtime",
"seed": 35,
"prompt": "Turn this into a watercolor painting",
"schedule_mu": 2.3,
"num_inference_steps": 3,
"enable_interpolation": false,
"output_feedback_strength": 1
}'
# Step 2: Poll result (replace <id>)
curl https://api.sandbase.ai/v1/run/<id> \
-H "Authorization: Bearer YOUR_API_KEY"API README
Flux 2 [klein] Realtime
Flux 2 Klein Realtime edits an existing image from a required text prompt. The required image anchors the visual material, while the prompt states the change, making the input relationship explicit from the start.
The request supports 1–8 inference steps, ten aspect ratios, seed control, interpolation, output-feedback strength from 0 to 1, and schedule-mu values from 0.3 to 2.5. These controls provide several ways to shape and repeat an edit while keeping image and prompt inputs required.
Highlights
Required image foundation. Every request begins with an image that supplies the visual material to revise.
Required written direction. A prompt accompanies the image and defines the requested edit.
Short inference range. Select 1–8 inference steps for the generation pass.
Realtime control set. Configure interpolation, output-feedback strength, schedule mu, aspect ratio, and seed for the intended run.
Pricing
| Billing unit | Price |
|---|---|
| Per request | $0.00194 |
When to Use
| ✅ Good fit | ❌ Consider alternatives |
|---|---|
| The model's named workflow matches the source material and intended output | A different input modality or model route is required |
| A managed asynchronous result is suitable for the production pipeline | A synchronous, interactive editor is essential |
| The documented controls cover the required duration, framing, or format | The project needs controls outside this endpoint's schema |
| Creative iteration benefits from a repeatable request structure | Exact deterministic pixels, frames, geometry, or samples are mandatory |
| A finished downloadable media asset is the desired deliverable | Editable source layers or a native project file are required |
Prompt Guide
For image-conditioned generation, state the intended result first, then add the subject or source treatment, progression, style, and delivery constraints. Keep one creative variable per phrase, use the documented field names for controls, and change one setting at a time when comparing results.
{
"aspect_ratio": "21:9",
"image": "https://example.com/start-frame.png",
"prompt": "Turn this into a watercolor painting"
}
Technical Specs
| Spec | Value |
|---|---|
| Model ID | bfl/flux-2/klein/realtime |
| Inputs | aspect_ratio, enable_interpolation, image, num_inference_steps, output_feedback_strength, prompt, schedule_mu, seed |
| Required inputs | prompt, image |
| Output fields | content_type, url |
| Execution | Async (submit, then poll for result) |
| Aspect Ratio | 21:9 / 16:9 / 3:2 / 4:3 / 5:4 / 1:1 / 4:5 / 3:4 / 2:3 / 9:16 |
Related Models
bfl/flux-2/klein/4b— Compare this concrete local family route.bfl/flux-2/klein/4b/base— Compare this concrete local family route.bfl/flux-2/klein/4b/base/edit— Compare this concrete local family route.