Flux2-Klein-9B-True-V2 is a text-to-image diffusion model hosted on Hugging Face. It serves users seeking to generate images from text prompts while operating under constrained hardware resources.
The model is distributed as a 9B parameter variant optimized for low VRAM environments. It runs in bfloat16 precision and can execute on consumer GPUs using only 4-6 GB of VRAM when paired with the recommended ComfyUI_UniBlockSwap plugin. Code examples demonstrate loading via the Diffusers library with torch_dtype set to bfloat16 and device_map set to cuda, or mps for Apple devices. The repository also lists compatibility with local applications including Draw Things and DiffusionBee.
It accepts natural language prompts such as "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" and produces corresponding images. The model card provides instructions for integration with notebooks on Google Colab and Kaggle in addition to direct library usage. Documentation appears in both English and Chinese.
The model carries an "other" license. It is made available for download and local inference at no cost through the Hugging Face platform.
Flux2 Klein 9B True sits in PulseGate's Image generation category. It focuses on generating high-quality images from text prompts on hardware with limited GPU memory. Flux2 Klein 9B True is an open-source project aimed at AI artists and developers. The project is open source (MIT). The product ships for the web, the command line, and API.
Behind Flux2 Klein 9B True is wikeeyang, and the product first shipped in 2026. The project is developed in the open on GitHub with 30 stars and 14 commits in the last 90 days. Among its 4 catalogued features are Text-to-Image Generation, Low VRAM Support, and Diffusers Compatible.
Latest indexed changes and source events
wikeeyang/Flux2-Klein-9B-True-V2 verified by the PulseGate indexer
