coder3101/gemma-3-4b-it-heretic
The coder3101/gemma-3-4b-it-heretic is a 4.3 billion parameter instruction-tuned multimodal language model, derived from Google's Gemma 3 family, with a 32K token context window. This version is specifically decensored using the Heretic v1.0.1 tool, significantly reducing refusals compared to the original Gemma 3-4b-it model. It excels in text generation, image understanding, and reasoning tasks, making it suitable for applications requiring less restrictive content filtering.
Loading preview...
Model Overview
The coder3101/gemma-3-4b-it-heretic is a 4.3 billion parameter instruction-tuned model based on Google's Gemma 3 family. It is a decensored variant of google/gemma-3-4b-it, created using the Heretic v1.0.1 tool. This modification significantly alters its refusal behavior, with reported refusals dropping from 98/100 in the original model to 1/100 in this version, while maintaining a low KL divergence of 0.26.
Key Capabilities
- Multimodal: Handles both text and image inputs, generating text outputs. Images are normalized to 896x896 resolution and encoded to 256 tokens.
- Extended Context: Features a 32K token context window for the 4B size, enabling processing of longer inputs.
- Multilingual Support: Supports over 140 languages.
- Reduced Refusals: Engineered to provide less restrictive responses compared to its base model.
- Versatile Applications: Suitable for text generation, image understanding, question answering, summarization, and reasoning tasks.
Ideal Use Cases
- Content Creation: Generating creative text formats, scripts, or marketing copy without strict content filters.
- Conversational AI: Powering chatbots and virtual assistants where broader response flexibility is desired.
- Research & Development: Experimenting with VLM and NLP techniques, especially in scenarios requiring less constrained model behavior.
- Image Analysis: Extracting and interpreting visual data for text communications.