v6.11.0
📦 invokeaiView on GitHub →
✨ 4 features🔧 4 symbols
Summary
This feature release introduces comprehensive support for the high-quality and fast FLUX.2 Klein (F2K) models, including new image editing capabilities. It also adds a Seed Variance Enhancer node for Z-Image Turbo to improve output diversity.
Migration Steps
- After updating, install the FLUX.2 Klein starter pack via the Model Manager.
- When using FLUX.2 Klein 4B (GGUF Q4), select the F2K VAE and the FLUX.2 Qwen3 text encoder in the Advanced section of the generation panel.
- When using FLUX.2 Klein 9B, select the F2K VAE and the FLUX.2 Klein Qwen3 8B Encoder in the Advanced section.
✨ New Features
- Added support for the new FLUX.2 Klein (F2K) image generation and edit models, including 4B and 9B variants and quantized versions.
- Implemented unique image editing feature for F2K allowing targeted modifications using descriptive prompts.
- Introduced a new FLUX.2 Klein starter pack in the Model Manager.
- Added Z-Image Turbo Seed Variance Enhancer node to increase image diversity by adding calibrated noise to prompt conditioning.