Model overview
What this model is for
Qwen3.8-Uncensored-NVFP4-MTP is a multimodal text-generation model mainly suited to coding and vision-related workflows. The Hugging Face listing identifies it as an NVFP4 and GGUF-oriented build with MTP and speculative-decoding support, and includes compatibility tags for llama.cpp and LM Studio. Its metadata also describes conversational use, text generation, and English and Chinese language coverage. The model is marked as uncensored and abliterated in its name and tags, which may make it relevant for applications requiring fewer built-in response restrictions, although the available source does not document specific safety behavior or evaluation results. Vision and multimodal support are listed, but the source does not provide details about supported image formats, resolution limits, or vision benchmarks. The underlying base model is identified as JonathanColetti/Qwen3.8-27B-Uncensored, while the exact parameter count for this quantized derivative is not disclosed. No inference-speed benchmarks, context-window specification, pricing, or detailed usage instructions are available in the supplied source snapshot. The listing carries an Apache-2.0 license tag; review the model card and files before deployment to confirm license applicability and runtime requirements.