Skip to content
View original post on X: Black Forest LabsOfficial· Pick60/100AI score60/100

Black Forest Labs releases open weights, quantized versions, and an API for its image model

AISummary

Black Forest Labs announces base and distilled open weights for its image model.

It also releases FP8, NVFP4, and INT8 quantized versions that run on GPUs and laptops with 8 GB of VRAM. The distilled model is available through the BFL API with native 4K output, bounding boxes, and pixel-perfect editing.

The company also lists launch support for AI Toolkit, ComfyUI, Diffusers, SGLang Diffusion, vLLM-Omni, and Nvidia fine-tuning recipes.

AIWhy it matters

The post lists exact checkpoint formats and hardware targets, showing which variants suit fine-tuning, fast generation, or laptop-scale self-hosting.

Post on XView on X
Black Forest LabsVerified on X
@bfl_ai

Part of a thread · earlier post

What we’re launching:

• Base open weights: the full checkpoint for fine-tuning, further training, and research.

• Distilled open weights: low-latency version for faster generation and self-hosting.

• Quantized versions: FP8, NVFP4 and INT8, small enough to run on GPUs and laptops with 8 GB of VRAM.

• BFL API: The distilled model with native 4K, bounding boxes, and pixel-perfect editing.

• Customization and deployment tools: Launch support for @ostrisai's AI Toolkit, @ComfyUI, Diffusers and SGLang Diffusion, vLLM-Omni, plus @nvidia fine-tuning recipes and deployment optimizations.

Source: Black Forest Labs · x.comPublished · added here

Topics

Image generationOpen-source ecosystemDeployment & engineeringModel releases