This is a LoRa extraction from FLUX.2 klein 9B. and FLUX.2 klein base 9B. I have created a series of LoRa use cases for the base model with 4–8 steps (CGF 1). extracted from: flux 2 klein base 9b fp8 and none base flux 2 klein base 9b bf16 and none base 3 Ranks for each 64/128/256 with option frobenius all bf16 in size and 3 standard loras bf16 is for bf16 model, fp8 for fp8 even if its bf16 in size with frobenious (experimental) I can create smaller models (e.g., Frobenius 075), but at the expense of quality... I see small differences — decide for yourself ;) extraction with LoraExtractKJ (using this parameters) no need for fp32 since base model is bf16 License This model falls under the FLUX Non Commercial License. Trademarks & IP This project may contain trademarks or logos for projects, products, or services. Use of Black Forest Labs and FLUX trademarks or logos in modified versions of this project must not cause confusion or imply sponsorship or endorsement. Any use of third party trademarks, intellectual property or logos are subject to those third party's policies.
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy