Black Forest Labs: FLUX.2 Klein 4B API

FLUX.2 [klein]-4B is the ultra-compact, high-speed variant of the FLUX.2 family, featuring a 4-billion parameter Rectified Flow Transformer. Designed for sub-second inference on consumer hardware, it can run on GPUs with as little as 13GB VRAM (e.g., RTX 3090/4070). Despite its small size, it unifies text-to-image generation, single-image editing, and multi-reference consistency into a single architecture. Released under the Apache 2.0 license, it is the most accessible frontier model for real-time interactive apps, edge deployment, and local development, delivering a visual quality-to-latency ratio that outperforms much larger models. The price is calculated based on the unit price per cost point, where the cost value is returned in the response after the task is submitted. For example: `json { "id": "5aaeac12-5dbc-43ff-93f0-eb61cfe7274b", "polling_url": "https://api.us2.bfl.ai/v1/get_result?id=5aaeac12-5dbc-43ff-93f0-eb61cfe7274b", "cost": 1.7, "input_mp": 3.0, "output_mp": 1.0 } `

Frequently Asked Questions

What is the knowledge cutoff of FLUX.2 Klein 4B?

The knowledge cutoff of FLUX.2 Klein 4B is 2025-09-30.