VibeThinker-3B-heretic

A decensored variant of WeiboAI/VibeThinker-3B (fine-tuned from Qwen/Qwen2.5-Coder-3B), produced with Heretic v1.4.0 (directional ablation / "abliteration"). Refusal behavior is suppressed via targeted weight edits to the attention output and MLP down-projections rather than fine-tuning, so the base model's thinking/reasoning gains and instruction-following are left largely intact.

Who this is for: users who want a ~3B reasoning/coding model with an explicit thinking mode (VibeThinker's chain-of-thought behavior) and no refusal guardrails - for CPU and low-VRAM local use, agents, and edge/on-device reasoning. Not a capability upgrade over base VibeThinker-3B - same model, refusal guardrails removed.

Runs on your gaming PC

Full GGUF ladder included — pick the quant that fits your card:

| Your GPU | Recommended quant | Weights |

| :--- | :--- | :--- |

| RTX 3090 / 4090 / 5090 (24 GB) | Q8_0 | ~3.3 GB |

| RTX 4080 / 5080 / 4060 Ti 16G (16 GB) | Q6_K | ~2.6 GB |

| RTX 3060 / 4070 / 5070 (12 GB) | Q5_K_M | ~2.3 GB |

| RTX 4060 / 3070 (8 GB) | Q4_K_M | ~2.0 GB |

| GTX 1660 Super / 2060 / 3050 laptop (6 GB) | IQ4_XS | ~1.9 GB |

| CPU-only / Apple Silicon | Q4_K_M | fits in system RAM |

Weights only, at this model's 3.1B native size; add ~1 GB for context.

OOM? Drop one quant level. Headroom to spare? Go one up.

Why abliteration instead of fine-tuning

Fine-tuning a "helpful" persona on top of RLHF'd refusals fights the base model's training and tends to degrade coherence. Abliteration instead finds and edits the specific weight directions responsible for refusal, leaving the rest of the network (and its capabilities) untouched. See the Heretic repo and the original abliteration writeup for the mechanism.

Files

GGUF quantizations

Full quantization set (14 quants + F16) produced with llama.cpp.

| File | Format | Size |

|---|---|---|

| VibeThinker-3B-heretic-F16.gguf | GGUF F16 | 5.75 GB |

| VibeThinker-3B-heretic-Q2_K.gguf | GGUF Q2_K | 1.19 GB |

| VibeThinker-3B-heretic-IQ3_S.gguf | GGUF IQ3_S | 1.36 GB |

| VibeThinker-3B-heretic-Q3_K_S.gguf | GGUF Q3_K_S | 1.35 GB |

| VibeThinker-3B-heretic-Q3_K_M.gguf | GGUF Q3_K_M | 1.48 GB |

| VibeThinker-3B-heretic-Q3_K_L.gguf | GGUF Q3_K_L | 1.59 GB |

| VibeThinker-3B-heretic-IQ4_XS.gguf | GGUF IQ4_XS | 1.63 GB |

| VibeThinker-3B-heretic-Q4_K_S.gguf | GGUF Q4_K_S | 1.71 GB |

| VibeThinker-3B-heretic-Q4_0.gguf | GGUF Q4_0 | 1.70 GB |

| VibeThinker-3B-heretic-Q4_1.gguf | GGUF Q4_1 | 1.86 GB |

| VibeThinker-3B-heretic-Q4_K_M.gguf | GGUF Q4_K_M | 1.80 GB |

| VibeThinker-3B-heretic-Q5_K_S.gguf | GGUF Q5_K_S | 2.02 GB |

| VibeThinker-3B-heretic-Q5_K_M.gguf | GGUF Q5_K_M | 2.07 GB |

| VibeThinker-3B-heretic-Q6_K.gguf | GGUF Q6_K | 2.36 GB |

| VibeThinker-3B-heretic-Q8_0.gguf | GGUF Q8_0 | 3.06 GB |

Qwen2.5-Coder architecture with VibeThinker thinking mode — loads natively in llama.cpp / Ollama / LM Studio / Jan.

Run llama serve -hf saidutta69/VibeThinker-3B-heretic to pull the default quant.

Quickstart

llama.cpp

# download the LTS build: https://github.com/ggerganov/llama.cpp/releases
llama-cli -m VibeThinker-3B-heretic-Q4_K_M.gguf -p "User: What is the fourth dimension?\nAssistant:" -n 512

Ollama

ollama run hf.co/saidutta69/VibeThinker-3B-heretic:Q4_K_M

LM Studio

1. Open LM Studio and click the search icon to open the Model Search panel.

2. Type "VibeThinker-3B-heretic" and click the download button marked GGUF.

3. Pick your quant, load the model, and start chatting.

Responsible use

This model has its refusal guardrails removed. Use it lawfully and ethically, and be aware that it may generate content the base model would refuse.

Made with ❤️ by RACER IS OP — follow for more uncensored models