Granite-4.2-3B-Heretic GGUF

GGUF quantizations of richardyoung/granite-4.2-3b-heretic, an abliterated build of ibm-granite/granite-4.2-3b (IBM's Granite 4.2 3B instruct model, multilingual (12 languages) with tool calling and a thinking mode) made with Heretic.

Abliteration Results

The very low KL divergence (0.085, far below the 0.5 “damage” threshold) means the model retains essentially all of its original capabilities and coherence.

Note on refusals: this build still refuses 27 of 100 harmful prompts, more than our other Heretic releases. That is deliberate: the trials that refused less also degraded the model's normal behavior (KL divergence above 0.10), and we chose to keep the original model's capabilities intact rather than trade them for fewer refusals.

Quick Start

With Ollama

ollama run richardyoung/granite-4.2-3b-heretic
# or straight from Hugging Face
ollama run hf.co/richardyoung/granite-4.2-3b-heretic-GGUF

With llama.cpp

huggingface-cli download richardyoung/granite-4.2-3b-heretic-GGUF \
    --include "*Q4_K_M*" --local-dir ./models

./llama-cli -m ./models/*Q4_K_M*.gguf -ngl 99

With Python (llama-cpp-python)

from llama_cpp import Llama

llm = Llama.from_pretrained(
    repo_id="richardyoung/granite-4.2-3b-heretic-GGUF",
    filename="*Q4_K_M*",
    n_gpu_layers=-1,
)

output = llm.create_chat_completion(
    messages=[{"role": "user", "content": "Explain abliteration in simple terms."}]
)
print(output["choices"][0]["message"]["content"])

Available Quantizations

Quant Quality

Each quant compared with the F16 model using llama-perplexity --kl-divergence on the wikitext-2 test set (40 × 512-token chunks). KL divergence measures how far the quant's next-token probabilities drift from F16; lower is better. Top-token match is how often the quant picks the same next token as F16. As a rough guide, a KLD under 0.01 is hard to tell from F16, and from about 0.05 differences start to show in longer outputs.

What is Abliteration?

Abliteration removes the "refusal direction" from a model's residual stream — a surgical modification that disables safety refusals without retraining. See Refusal in Language Models Is Mediated by a Single Direction. Heretic automates it: it searches ablation parameters with an optimizer (200 trials here) and keeps the trials that best trade off fewer refusals against KL divergence from the original model.

Intended Use

Research, creative writing, education on alignment techniques, and unrestricted local inference. This model will comply with requests the original refuses; use responsibly and in accordance with applicable laws. License: Apache-2.0, inherited from the base model.

Other Models by richardyoung

• New Heretic models (reproducible, listed in the Heretic Grimoire): LFM2.5-1.2B | Granite-4.2-3B | Granite-4.2-8B | Olmo-3-7B-Think | Phi-4-mini | DeepSeek-R1-0528-Qwen3-8B | Agents-A1-4B | FrogNano-4B | Muse-Glimmer-30B | Hemmingway-1
• Abliterated/Uncensored models: Qwen2.5-7B | Qwen3-14B | DeepSeek-R1-32B | Qwen3-8B
• OCR & Vision: MiniCPM-V-4.6 GGUF | dots.mocr GGUF | Unlimited-OCR GGUF | olmOCR GGUF | LightOnOCR-3-4B GGUF | WeVisDoc-4B GGUF | Xiaomi-OCR-0 GGUF
• MLX quantizations (Apple Silicon): Kimi-K2 series | olmOCR MLX
• Healthcare/Medical: Synthea 575K patients dataset | CardioEmbed
• Research: LLM Instruction-Following Evaluation (arxiv:2510.18892)

---

Built & maintained by Richard Young · DeepNeuro