← back to catalog · registered 2026-11-14 09:03

prithivMLmods/Qwen3.5-9B-DS-v4-Flash-uncensored-heretic

prithivMLmods Qwen 9B multimodal
Your rig guess connected
? Why do I need an app?
Reading your rig…

This is a rough estimate. Install the free app - we'll show exact numbers.

Reading real hardware from your app right now. Numbers below are exact.

Below is the per-quantization compatibility for this model.

curl -H "Authorization: Bearer $ABL_KEY" \
     "https://abliteration.org/api/v1/models/prithivMLmods%2FQwen3.5-9B-DS-v4-Flash-uncensored-heretic"
Response includes
  • classification m3
  • files 13
  • benchmarks 11 entries
  • author_summary 117 models
  • readme_text full
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M3
Primary method

Layer-wise ablation

Applied on top of direct removal inherited from the base model.
Confidence
HIGH
Inherited from base model
Why this label 2 signals
Producer identity confirmed by naming conventions, tags or the model card. This label is very unlikely to change.
  • 'heretic' in model name (Heretic-produced)
  • Heretic uses layer-wise optimization (M3) with underlying direction removal (M1)
Refusal direction extracted via
Extraction technique

Difference-of-means

Confidence
HIGH
Why we say so
name contains 'heretic'; Heretic default extraction is difference-of-means (Arditi 2024)
Downloads · 30-day
78
Likes
2
Descendants
4
in 4 direct forks
Model age
1w ago
created 2026-11-13
Downloads over time
Now 78 → from 0 ↑ 0%
0295786 0 on Nov 14 78 on Nov 22 Nov
Nov 14 → Nov 22 · 9 snapshots · spans 8 days

Benchmarks

Portrait before abliteration
Benchmarks of the base model as it stood before the refusal-removal operation. Compare with the numbers above to see what the operation cost.
Benchmark Score Source
Entertainment 1.4 UGI
Hazardous 2.4 UGI
Natural Intelligence 17.62 UGI
Political lean -12.2% UGI
Sensitive-Info 14.65 UGI
SocPol 0.9 UGI
UGI 17.27 UGI
Willingness (10) 2.2 UGI
W10-Adherence 1.5 UGI
W10-Direct 3 UGI
Writing 33.52 UGI

Genealogy 4 direct forks

Full fork graph →

This model's place in the market. Above: what it was derived from. Below: the tree of everything derived from it.

Variants by this author 4 formats · 3K downloads combined

The same weights this author released in different packaging. Pick the format that matches your runtime.

Metadata

License
apache-2.0
Languages
en
Tags
transformers safetensors qwen3_5 image-text-to-text mtp text-generation-inference llama-cpp Qwen3.5 flash math reasoning sft

Related

Total size
18.0 GB
Files
13
Quantizations
1
Registered
2026-11-14 09:03
Last updated on HF
2026-11-14 07:11

Files by quantization

Auxiliary files 13 files 18.0 GB
model-00003-of-00004.safetensors 4.83 GB 2d96a777 download
model-00002-of-00004.safetensors 4.83 GB 965e300b download
model-00001-of-00004.safetensors 4.82 GB 153ad950 download
model-00004-of-00004.safetensors 3.49 GB 63202c56 download
tokenizer.json 19.1 MB 9de52859 download
model.safetensors.index.json 68.7 KB 336eee7a download
chat_template.jinja 7.57 KB a585dec8 download
README.md 5.60 KB dff4e4bf download
config.json 2.87 KB b99eb4a2 download
.gitattributes 1.53 KB 52373fe2 download
tokenizer_config.json 1.30 KB 6d2df0d3 download
processor_config.json 1.29 KB a9804dd9 download
generation_config.json 153 B f625b634 download

README current version from Hugging Face


tags:

  • mtp
  • text-generation-inference
  • llama-cpp
  • Qwen3.5
  • flash
  • math
  • reasoning
  • sft
  • deepseek-v4
  • bf16
  • v3.0
  • abliterated
  • heretic
  • uncensored
  • decensored
  • instruction-following
    license: apache-2.0
    language:
  • en
    base_model:
  • Qwen/Qwen3.5-9B
    pipeline_tag: image-text-to-text
    library_name: transformers

1

Qwen3.5-9B-DS-v4-Flash-uncensored-heretic

Qwen3.5-9B-DS-v4-Flash-uncensored-heretic is a fine-tuned (trained on DeepSeek V4 Flash reasoning traces) and uncensored version of Qwen/Qwen3.5-9B, created using Heretic v2.0.0.dev0. The model combines reasoning-oriented fine-tuning with Heretic-based uncensoring to explore improvements in long-form reasoning, mathematical problem solving, scientific analysis, coding, and instruction following while reducing refusal behavior.

[!NOTE]
MTP — Speculative Decoding: Yes | This model is an experimental release and may generate unexpected behaviors or reasoning artifacts in certain scenarios. Its uncensored behavior does not guarantee factual accuracy, and outputs should be independently evaluated.

Performance

This is a fine-tuned (trained on DeepSeek V4 Flash reasoning traces) and uncensored version of Qwen/Qwen3.5-9B, created using Heretic v2.0.0.dev0.

Metric This Model Original Model (Qwen/Qwen3.5-9B)
Refusals 5/100 100/100
KL Divergence 0.0825 0 (by definition)

Key Highlights

  • Qwen 3.5 Foundation: Based on the Qwen/Qwen3.5-9B foundation model.
  • Reasoning Fine-Tuning: Trained using DeepSeek V4 Flash reasoning traces to enhance multi-step reasoning and analytical capabilities.
  • Heretic Uncensoring: Created using Heretic v2.0.0.dev0 to reduce model refusal behavior.
  • Long-Form Reasoning: Focuses on complex analytical tasks, mathematical problem solving, scientific reasoning, and coding.
  • Instruction Following: Designed to improve responsiveness across a variety of user instructions.
  • Research-Focused Release: Intended for experimentation, model evaluation, and reasoning research.
  • Efficient 9B Deployment: Suitable for local inference and research environments.

Quick Start with Transformers

Install dependencies

pip install -U transformers accelerate torch

Load the model

from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

model_id = "prithivMLmods/Qwen3.5-9B-DS-v4-Flash-uncensored-heretic"

tokenizer = AutoTokenizer.from_pretrained(model_id)

model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype="auto",
    device_map="auto"
)

messages = [
    {
        "role": "user",
        "content": "Explain how a transformer model processes text."
    }
]

inputs = tokenizer.apply_chat_template(
    messages,
    tokenize=True,
    add_generation_prompt=True,
    return_tensors="pt"
).to(model.device)

outputs = model.generate(
    inputs,
    max_new_tokens=512,
    temperature=0.7,
    top_p=0.9,
    do_sample=True
)

print(
    tokenizer.decode(
        outputs[0][inputs.shape[-1]:],
        skip_special_tokens=True
    )
)

Intended Use

  • Reasoning Research: Exploring reasoning-oriented fine-tuning and uncensoring techniques.
  • Mathematical Reasoning: Evaluating multi-step mathematical problem solving.
  • Scientific Reasoning: Performing structured scientific analysis and problem solving.
  • Coding Assistance: Supporting code understanding, generation, and debugging.
  • Instruction Following: Evaluating instruction adherence and response generation.
  • Uncensored Model Research: Studying refusal behavior and the effects of uncensoring on model responses.
  • Local Deployment: Running an experimental 9B language model in local inference environments.

Limitations

  • Experimental Model: Behavior may differ from the original Qwen3.5-9B model.
  • Reasoning Artifacts: Complex reasoning chains may occasionally contain incorrect intermediate steps or conclusions.
  • Factual Reliability: Fine-tuning does not guarantee factual accuracy or eliminate hallucinations.
  • Uncensored Outputs: Reduced refusal behavior may result in responses to requests that the original model would decline.
  • Training Biases: Performance reflects the characteristics and coverage of the training data.
  • Safety Considerations: Outputs should be independently evaluated before being used in production or safety-critical applications.

Acknowledgements

  • Qwen/Qwen3.5-9B: Original foundation model used for this project.

  • DeepSeek V4 Flash: Reasoning traces used for reasoning-oriented fine-tuning.

  • Heretic: Automated model uncensoring framework used to create this release. Version: v2.0.0.dev0.

  • TRL - Transformers Reinforcement Learning: Training library providing tools for supervised fine-tuning, preference optimization, reward modeling, and other language model training methods.

  • Transformers: Hugging Face library providing tools for loading, training, and running state-of-the-art machine learning models.

README history 11 versions

The author's README evolved over time. Click a version to see its content at that point.

  1. 2026-11-14 Update README.md 11120f0 5.6 KB
    Loading...
  2. 2026-11-14 Update README.md 417c4ae 5.6 KB
    Loading...
  3. 2026-11-14 Update README.md c77b336 5.6 KB
    Loading...
  4. 2026-11-13 Update README.md c76e5d2 5.5 KB
    Loading...
  5. 2026-11-13 Update README.md d6d21dc 5.4 KB
    Loading...
  6. 2026-11-13 Update README.md 59824b1 5.4 KB
    Loading...
  7. 2026-11-13 Update README.md ace8a97 5.4 KB
    Loading...
  8. 2026-11-13 Update README.md f99c367 747 B
    Loading...
  9. 2026-11-13 Update README.md a4ffb82 526 B
    Loading...
  10. 2026-11-13 Update README.md c67c765 295 B
    Loading...
  11. 2026-11-13 Create README.md d0240ef 289 B
    Loading...
Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.

Open in Abliteration