AI BriefingDeepSeek-V4-Pro-Qwen3.5-9B-UNCENSORED-heretic-i1-GGUF: 9B DeepSeek-V4-Pro, uncensored version
mradermacher/DeepSeek-V4-Pro-Qwen3.5-9B-UNCENSORED-heretic-i1-GGUF
About the project
A GGUF quantized model restructured to the Qwen3.5 9B scale based on the DeepSeek-V4-Pro architecture. It operates with the safety filters or censorship mechanisms of the original model removed, as indicated by the 'UNCENSORED-heretic' tag.
It is ready to use with major local inference frameworks such as llama.cpp, Ollama, and LM Studio. It offers various quantization levels from IQ1_M to Q6_K, allowing you to adjust memory usage to fit your hardware specifications.
It supports reasoning, mathematics, STEM, tool-use, and function-calling capabilities. It handles 6 languages and is distributed under the Apache-2.0 license, enabling commercial use.
It is suitable for development environments requiring uncensored free text generation or situations where a lightweight LLM must be run locally. imatrix-based quantization minimizes performance degradation compared to the original model.
mradermacher/DeepSeek-V4-Pro-Qwen3.5-9B-UNCENSORED-heretic-i1-GGUF
The original page has no description.
This introduction was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report errors, attribution issues, or removal requests via Contact.