GEMMA 4 E4B ABLITERATEDA ComfyUI-compatible repack of wangzhang's abliterated Gemma 4 E4B weights, formatted as a single safetensors file usable with ComfyUI's standard Load CLIP and TextGenerate nodes.WHAT THIS ISThis file combines:- The abliterated language model weights from wangzhang/gemma-4-E4B-it-abliterix <a target="_blank" rel="ugc" href="https://huggingface.co/wangzhang/gemma-4-E4B-it-abliterix">https://huggingface.co/wangzhang/gemma-4-E4B-it-abliterix</a>- The vision encoder, audio encoder, multimodal projector, and tokenizer components from Comfy-Org/gemma-4 (unmodified) <a target="_blank" rel="ugc" href="https://huggingface.co/Comfy-Org/gemma-4">https://huggingface.co/Comfy-Org/gemma-4</a>The result is a drop-in replacement for ComfyUI's standard Gemma 4 E4B text encoder file, with the language model behavior of the upstream abliterated weights.WHAT THIS IS NOTThis is not a new model. The intelligence, capabilities, and modifications were done upstream by Google (base model) and wangzhang (abliteration). This file is a format repack it puts existing modified weights into ComfyUI's expected single-file structure so users can load them with standard nodes.CREDITSBase model: Google DeepMind — Gemma 4 E4B IT<a target="_blank" rel="ugc" href="https://huggingface.co/google/gemma-4-E4B-it">https://huggingface.co/google/gemma-4-E4B-it</a>Abliteration work: wangzhang — Abliterix method<a target="_blank" rel="ugc" href="https://huggingface.co/wangzhang/gemma-4-E4B-it-abliterix">https://huggingface.co/wangzhang/gemma-4-E4B-it-abliterix</a>ComfyUI integration and reference format: Comfy-Org / Kijai<a target="_blank" rel="ugc" href="https://huggingface.co/Comfy-Org/gemma-4">https://huggingface.co/Comfy-Org/gemma-4</a>INSTALLATIONPlace the .safetensors file in:ComfyUI/models/text_encoders/Use with ComfyUI's standard Load CLIP and TextGenerate nodes by selecting this file as the text encoder.TECHNICAL NOTESThe repack reconstructs Gemma 4's shared-KV layer structure: layers 24-41 share key and value projections with layer 23, requiring tensor cloning during conversion. The k_norm parameters for these shared layers are sourced from the reference Comfy-Org file to maintain compatibility with ComfyUI's loader.Vision encoder (4 layers, 768 hidden), audio encoder (12 layers), and multimodal projector are bundled unchanged from the reference file. Tokenizer is the standard Gemma 4 tokenizer.LICENSEThis model is provided under the Gemma Terms of Use:<a target="_blank" rel="ugc" href="https://ai.google.dev/gemma/terms">https://ai.google.dev/gemma/terms</a>Use of this model is also subject to Google's Gemma Prohibited Use Policy:<a target="_blank" rel="ugc" href="https://ai.google.dev/gemma/prohibited_use_policy">https://ai.google.dev/gemma/prohibited_use_policy</a>By downloading this model, you agree to be bound by these terms. You are responsible for your use of the model and any content it generates.RESPONSIBLE USEThis is an abliterated model with reduced refusal behavior on the base language model. The vision and audio encoders are unmodified.The user is responsible for ensuring their use complies with applicable laws and the Gemma Prohibited Use Policy. The author of this repack does not endorse generating content that would violate those terms.

Files with SHA256: ba756436a278e5c683535a432b12425278d365274cfb303220f6e7f86d7d92af