---

license: apache-2.0

tags:

• quantized
• gguf
• abliterated
• qwen3
• moe

base_model:

• Qwen/Qwen3-30B-A3B
• Qwen/Qwen3.6-35B-A3B

---

Qwen3.6-35B-A3B-Abliterated-Heretic GGUF

<!-- polished-overview:start -->

Overview

Qwen3.6-35B-A3B-Abliterated-Heretic-GGUF is a GGUF release for llama.cpp-compatible runtimes and local inference, published by groxaxo.

It is intended for open-source evaluation, reproducible experimentation, and compatible local or

hosted inference workflows. The wording below is deliberately limited to what can be verified

from this repository's metadata and artifacts.

The repository name identifies a behavior-modified or reduced-filtering lineage. That label describes the source or conversion history; it is not a guarantee of unrestricted behavior in every prompt or runtime. Test outputs carefully before sharing or deploying them.

At a glance

| Field | Details |

|---|---|

| Format | GGUF |

| Source / base | Qwen/Qwen3-30B-A3B |

| Intended task | image-text-to-text |

| License | apache-2.0 |

What is included

• *.gguf (10 files)
• Additional configuration, tokenizer, processor, or shard files (10 visible artifacts total)

Quick start

llama.cpp

Download a .gguf file that fits your available memory, then run it with a current llama.cpp

build:

llama-cli \
  -m /path/to/model.gguf \
  -p "Write a concise technical summary."

For vision or any-to-any models, download the matching multimodal projection file when one is

provided and follow the source model's modality-specific instructions.

Compatibility and responsible use

• Use a runtime that explicitly supports this format, architecture, and modality.
• Keep configuration, tokenizer, processor, projection, and weight files from the same revision together.
• Review the source model card and license before redistribution or deployment.
• Hardware needs depend on parameter count, context length, cache precision, quantization, and concurrency.
• Report reproducible issues with the runtime version, hardware, launch command, and a minimal example.

Quantization or conversion changes numerical behavior, memory use, and throughput relative to the source checkpoint; validate quality on your own workload.

Generated outputs may be inaccurate or unsuitable for a given use case. Users are responsible for

testing behavior, applying appropriate safeguards, and complying with applicable licenses and laws.

<!-- polished-overview:end -->

Quantized GGUF files for Qwen3.6-35B-A3B-Abliterated-Heretic.

Available Quantizations

| File | Quant | Size | Notes |

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

| Qwen3.6-35B-A3B-Abliterated-Heretic-Q8_0.gguf | Q8_0 | ~35 GB | Best quality, largest |

| Qwen3.6-35B-A3B-Abliterated-Heretic-Q6_K.gguf | Q6_K | ~27 GB | High quality |

| Qwen3.6-35B-A3B-Abliterated-Heretic-Q5_K_M.gguf | Q5_K_M | ~24 GB | Good quality/size balance |

| Qwen3.6-35B-A3B-Abliterated-Heretic-Q4_K_M.gguf | Q4_K_M | ~20 GB | Recommended |

| Qwen3.6-35B-A3B-Abliterated-Heretic-Q4_K_S.gguf | Q4_K_S | ~19 GB | Smaller Q4 variant |

| Qwen3.6-35B-A3B-Abliterated-Heretic-Q3_K_M.gguf | Q3_K_M | ~16 GB | Lower bitrate |

| Qwen3.6-35B-A3B-Abliterated-Heretic-IQ4_XS.gguf | IQ4_XS | ~18 GB | Importance-matrix quant |

| Qwen3.6-35B-A3B-Abliterated-Heretic-Q2_K.gguf | Q2_K | ~13 GB | Smallest, lowest quality |

| mmproj-BF16.gguf | BF16 | ~861 MB | Multimodal projection |

Source

• Base model: Qwen/Qwen3.6-35B-A3B
• Abliterated model: groxaxo/Qwen3.6-35B-A3B-Abliterated-Heretic

Usage

Use with llama.cpp, LM Studio, Ollama, or any GGUF-compatible inference engine.

# Example with llama.cpp
./llama-server -m Qwen3.6-35B-A3B-Abliterated-Heretic-Q4_K_M.gguf --mmproj mmproj-BF16.gguf -ngl 99

Model Details

Qwen3.6-35B-A3B is a Mixture-of-Experts model with 256 experts (8 active), totaling ~35B parameters but only ~3B active per token. Features SSM (State Space Model) layers alongside attention.

This "Abliterated-Heretic" version has had alignment/refusal training removed via abliteration techniques.