license: other license_name: nvidia-open-model-license license_link: https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-license/ tags:

• uncensored
• nemotron
• mamba2
• hybrid language:
• en base_model: nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16

Nemotron3-Nano-4B-Uncensored-HauhauCS-Aggressive

Join the Discord for updates, roadmaps, projects, or just to chat.

NVIDIA Nemotron-3 Nano 4B uncensored by HauhauCS. 0/465 refusals.

HuggingFace's "Hardware Compatibility" widget doesn't recognize K_P quants — it may show fewer files than actually exist. Click "View +X variants" or go to Files and versions to see all available downloads.

About

No changes to datasets or capabilities. Fully functional, 100% of what the original authors intended - just without the refusals.

These are meant to be the best lossless uncensored models out there.

This release has NVIDIA's GenRM (generative reward model) fully removed. GenRM acts as an internal critic that scores and filters the model's own outputs — effectively a second layer of censorship on top of the base refusals. Removing it gives you the raw model output without any self-censoring.

For a comparison build with GenRM still active, see Nemotron3-Nano-4B-Uncensored-HauhauCS-Aggressive-GenRM (IQ2_M only, for side-by-side testing).

Aggressive Variant

Stronger uncensoring — model is fully unlocked and won't refuse prompts. May occasionally append short disclaimers (baked into base model training, not refusals) but full content is always generated.

For a more conservative uncensor that keeps some safety guardrails, check the Balanced variant when it's available.

Downloads

All quants generated with importance matrix (imatrix) for optimal quality preservation on abliterated weights.

What are K_P quants?

K_P ("Perfect") quants are HauhauCS custom quantizations that use model-specific analysis to selectively preserve quality where it matters most. Each model gets its own optimized quantization profile.

A K_P quant effectively bumps quality up by 1-2 quant levels at only ~5-15% larger file size than the base quant. Fully compatible with llama.cpp, LM Studio, and any GGUF-compatible runtime — no special builds needed.

Note: K_P quants may show as "?" in LM Studio's quant column. This is a display issue only — the model loads and runs fine.

Specs

• 3.97B parameters
• Hybrid Mamba2-Transformer architecture (42 layers: 21 Mamba2, 17 MLP, 4 Attention)
• 262K native context
• Thinking/reasoning mode (toggleable)
• Tool calling support
• Compressed from NVIDIA Nemotron-Nano-9B-v2 using Nemotron Elastic framework
• Based on nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16

Recommended Settings

From the official NVIDIA authors:

Reasoning mode (default, thinking enabled):

• temperature=1.0, top_p=0.95

Tool calling:

• temperature=0.6, top_p=0.95

Disabling reasoning:

• Set enable_thinking=False in chat template — trades accuracy for speed on simpler tasks

Usage

Works with llama.cpp, LM Studio, Jan, koboldcpp, and other GGUF-compatible runtimes.

llama-cli -m Nemotron3-Nano-4B-Uncensored-HauhauCS-Aggressive-Q4_K_P.gguf \
  --jinja -c 131072 -ngl 99

Note: LM Studio may show unexpected values in architecture/params columns — this is a display quirk with hybrid Mamba models, the model runs correctly.

Other Versions

• Nemotron3-Nano-4B-Uncensored-HauhauCS-Aggressive-GenRM — same abliteration but with GenRM still active (IQ2_M only, for comparison)