Primary use stated or inferred cautiously from official documentation.
F
BLACK FOREST LABS / black-forest-labs/flux-3-image-klein-base
FLUX 3 Image Klein Base
A 5.6-billion-parameter image model released on October 15, 2026. It generates and edits images from text and references, with bounding-box composition control. Distilled and Base are different checkpoints; neither should be confused with the full FLUX 3 Image model.
01 / SUMMARY
The essential
Maximum input capacity when published by the source.
Documented maximum generation limit.
Verified availability channels.
Primary functional family and verified specialties.
Declared terms for API access, model weights, or self-hosting.
02 / TECHNICAL SHEET
Limits and integration
| API ID | black-forest-labs/flux-3-image-klein-base |
|---|---|
| Model type | Image |
| Access model | Restricted weights |
| License | FLUX Kommunity License |
| Deployment | Self-hosted |
| Release | 15 OCT 2026 |
| Knowledge cutoff | Not published |
| Entrance | Text · Image |
| Exit | Image |
| Context window | Up to ten reference images |
| maximum output | Up to 4K |
| Reasoning | Not published |
| Published tools | |
| Structured Outings | Not published |
| Batch processing | Not published |
| Prompt cache | Not published |
| Fine-tuning | Not published |
| Verified platforms | Hugging Face · Diffusers · ComfyUI |
Local visual creation and editing with up to ten references and up to 4K output. Distilled uses 16 steps and CFG 1; Base uses 50 steps and CFG 3.5 and targets fine-tuning and LoRA training.
Use the official prompt upsampler or full captions: the Diffusers pipeline does not run it automatically. The upsampler outputs text, not images. Weights use the FLUX Kommunity License, not Apache: commercial rights depend on the requirements and thresholds in the license.
03 / CAPABILITIES
What it can do
Orientation
Local visual creation and editing with up to ten references and up to 4K output. Distilled uses 16 steps and CFG 1; Base uses 50 steps and CFG 3.5 and targets fine-tuning and LoRA training.
Context
Up to ten reference images · Up to 4K
Tools and integration
Not published
Access
Hugging Face
Input modalities
TextImage
04 / PRICES
Documented cost
| Concept | Worth | Unit/condition |
|---|---|---|
| Standard input | No per-token price | Standard API rate |
| Cached input | Not published | Reading reused prefixes |
| Cache write or storage | Not published | The condition varies by provider |
| Standard output | Self-hosted infrastructure cost | May include reasoning tokens |
| Batch input | Not published | Asynchronous processing |
| Batch output | Not published | Asynchronous processing |
The Distilled API charges per image: USD 0.008 at 768sq; 0.009 at 1K; 0.011 at 1.5K; 0.014 at 2K and 0.036 at 4K. No commercial Base endpoint has been verified. These are not token tariffs or local execution costs.
i Prices change and may depend on level, region or context length. Check the source before making a decision.
05 / EVALUATIONS
How to read the results
“
No numerical scores specific to Klein have been verified. We do not inherit FLUX 3 Image benchmarks. Inferama has not evaluated it yet; record checkpoint, upsampler, resolution, references, seed, steps and hardware to measure fidelity, cost and latency.
i No verified scores; Inferama evaluation pending.
Inferama only highlights a “best result” when the metric, test set, configuration, and date allow for an equivalent comparison. The supplier's figures are presented as claims from its own source.
06 / VERSIONS
Chronology
Klein Base
Local visual creation and editing with up to ten references and up to 4K output. Distilled uses 16 steps and CFG 1; Base uses 50 steps and CFG 3.5 and targets fine-tuning and LoRA training.
07 / SOURCES