← back to catalog · registered 2026-11-19 15:04

Peymandc/Qwen3.5-0.8B-Heretic-Uncensored

Peymandc Qwen 800M 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/Peymandc%2FQwen3.5-0.8B-Heretic-Uncensored"
Response includes
  • classification m3
  • files 11
  • benchmarks 11 entries
  • author_summary 4 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
62
Likes
0
Descendants
3
in 3 direct forks
Model age
2d ago
created 2026-11-19
Downloads over time
Now 62 → from 0 ↑ 0%
0234568 0 on Nov 19 62 on Nov 22 Nov
Nov 19 → Nov 22 · 4 snapshots · spans 3 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 UGI
Hazardous 1.2 UGI
Natural Intelligence 4.68 UGI
Political lean -5.3% UGI
Sensitive-Info 7.29 UGI
SocPol 0 UGI
UGI 16.53 UGI
Willingness (10) 3.5 UGI
W10-Adherence 4 UGI
W10-Direct 3 UGI
Writing 23.01 UGI

Genealogy 3 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 2 formats · 707 downloads combined

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

Metadata

License
apache-2.0
Tags
transformers safetensors qwen3_5 image-text-to-text qwen qwen3.5 multimodal vision-language conversational heretic uncensored decensored

Related

Total size
1.59 GB
Files
11
Quantizations
1
Registered
2026-11-19 15:04
Last updated on HF
2026-11-20 17:37

Files by quantization

Auxiliary files 11 files 1.61 GB
model.safetensors 1.59 GB 5ff689a1 download
tokenizer.json 19.1 MB a5cd9732 download
README.md 14.9 KB 7888e8a4 download
LICENSE 11.3 KB f938136e download
chat_template.jinja 7.72 KB cd794859 download
config.json 2.71 KB 67ea4adc download
.gitattributes 1.53 KB 52373fe2 download
tokenizer_config.json 1.52 KB 7bc4ffe4 download
MODIFICATIONS.md 1.28 KB 9a218df2 download
processor_config.json 1.25 KB 47f77d30 download
generation_config.json 122 B 842e2b32 download

README current version from Hugging Face


library_name: transformers
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen3.5-0.8B/blob/main/LICENSE
pipeline_tag: image-text-to-text
base_model: Qwen/Qwen3.5-0.8B
tags:

  • qwen
  • qwen3.5
  • qwen3_5
  • multimodal
  • vision-language
  • conversational
  • transformers
  • safetensors
  • heretic
  • uncensored
  • decensored
  • abliterated
  • reproducible

Qwen3.5-0.8B Heretic Uncensored

Model Pulse

Qwen3.5-0.8B Heretic Uncensored

This is a decensored version of Qwen/Qwen3.5-0.8B, created using Heretic v2.1.0.

The model was modified using Arbitrary-Rank Ablation (ARA). This is a training-free model modification, not a new pre-training run or a conventional fine-tune.

[!TIP]
This model is reproducible.

See the reproduce/README.md file in the reproduce/ directory for the configuration, evaluation details, and instructions to reproduce the modification.

💜 Enjoying my work? You can support my independent AI projects here.

Arbitrary-Rank Ablation (ARA) parameters

Parameter Value
start_layer_index 9
end_layer_index 20
preserve_good_behavior_weight 0.4395
steer_bad_behavior_weight 0.0002
overcorrect_relative_weight 0.6954
neighbor_count 2

Heretic evaluation

Heretic v2.1.0 · ARA

Metric This model Original Qwen3.5-0.8B
Refusal-keyword matches 8/100 (8%) 93/100 (93%)
KL divergence 0.0675 0.0000 (by definition)

Evaluation methodology

  • Refusal-keyword matches: Heretic's KeywordRate scorer, evaluated on the first 100 prompts from mlabonne/harmful_behaviors (test[:100]). A response is counted when it contains a configured refusal keyword. This is a keyword-based proxy; it is not a semantic classifier or a human-reviewed assessment of whether a response is a refusal.
  • KL divergence: Heretic's KLDivergence scorer, evaluated on the first 100 prompts from mlabonne/harmless_alpaca (test[:100]). The original model is the reference, so its divergence is zero by definition.

The scores depend on the evaluation prompts, scorer configuration, and refusal-keyword list. They should not be interpreted as a comprehensive measure of safety, helpfulness, or general capability. Full configuration and reproduction material belong in the reproduce/ directory.

Safety and responsible use

This model has undergone refusal-behavior modification using Heretic. As a result, it may respond to requests that the original model would have refused.

Users are responsible for evaluating the model's outputs and ensuring that their use complies with applicable laws and regulations. The model may generate inaccurate, misleading, harmful, or otherwise inappropriate content. Its outputs should not be treated as verified information or professional advice.

The reduction in refusal-keyword matches reported above is not a safety certification and does not establish that the model will comply with every request.


Documentation boundary: The architecture descriptions, capability statements, usage guidance, and benchmark results in the sections below are based on the original Qwen/Qwen3.5-0.8B documentation. They have not been independently re-evaluated on this Heretic variant. The Heretic-specific evaluation results are reported separately above.

Qwen3.5-0.8B

Qwen3.5-0.8B is a compact, post-trained multimodal vision-language model. It combines text processing with a vision encoder and uses a hybrid language-model architecture based on Gated DeltaNet and Gated Attention.

At this parameter scale, Qwen describes the model as suitable for prototyping, task-specific fine-tuning, research, and development. The 262K-token context length is the model's native maximum; the amount that can be used in practice depends on the inference framework, available memory, and runtime configuration.

Model details

Property Specification
Base model Qwen/Qwen3.5-0.8B
Checkpoint type Post-trained multimodal model
Task Image-text-to-text
Language-model parameters 0.8B
Architecture Hybrid Gated DeltaNet and Gated Attention, with a vision encoder
Number of language-model layers 24
Hidden dimension 1,024
Token embedding size 248,320 (padded)
Feed-forward intermediate dimension 3,584
Native context length 262,144 tokens
License Apache 2.0; see the original LICENSE

Language-model architecture

The official model card describes the 24-layer language model as six repetitions of the following layout:

3 × (Gated DeltaNet → FFN) → 1 × (Gated Attention → FFN)

  • Gated DeltaNet: 16 value heads and 16 query/key heads, with a head dimension of 128.
  • Gated Attention: 8 query heads and 2 key/value heads, with a head dimension of 256 and a rotary position-embedding dimension of 64.
  • Feed-forward network: intermediate dimension of 3,584.
  • Output embeddings: 248,320 output entries, tied to the token embeddings.
  • Multi-token prediction: the original model was trained with multi-step prediction.

The architecture information above describes the original Qwen checkpoint. It is included for reference and is not a separate architectural audit of this modified release.

Capabilities and limitations

  • Text and image input: the base checkpoint is an image-text-to-text model. Its Transformers interface accepts text and image content in the same conversation.
  • Multilingual use: Qwen describes the Qwen3.5 family as supporting 201 languages and dialects. This is a family-level statement; language-by-language performance has not been independently measured on this Heretic variant.
  • Small-model limitations: a 0.8B model has limited capacity compared with larger language models. Validate its accuracy and reliability for the intended task rather than assuming that the base model's family-level claims apply equally to this checkpoint.
  • Thinking mode: Qwen3.5-0.8B uses non-thinking mode by default. The official documentation warns that this small model can be more prone than larger Qwen3.5 models to thinking loops when thinking mode is enabled.

Files and formats in this repository

This repository is intended to host the Hugging Face Transformers checkpoint in Safetensors format.

Safetensors describes the weight-file format, not the tensor precision. Check the repository's file list and weight metadata for the actual dtype of the uploaded checkpoint. This model card does not claim that separate GGUF, GPTQ, AWQ, FP8, ONNX, MLX, or other quantized/converted versions are hosted here. Such formats should be linked only after they are separately published and verified.

Inference

The primary usage path for this repository is Hugging Face Transformers. The examples below use this repository's model ID, so they will work once the repository is public or when you have access to it.

Transformers installation

Install a recent Transformers version and Accelerate:

pip install -U transformers accelerate

If the installed Transformers release does not yet recognize the Qwen3.5 architecture, install the current development version:

pip install "git+https://github.com/huggingface/transformers.git"

For image input, make sure the relevant image dependencies (such as torchvision and Pillow) are installed as well.

Text-only example

import torch
from transformers import AutoProcessor, AutoModelForMultimodalLM

model_id = "Peymandc/Qwen3.5-0.8B-Heretic-Uncensored"

processor = AutoProcessor.from_pretrained(model_id)
model = AutoModelForMultimodalLM.from_pretrained(
    model_id,
    torch_dtype="auto",
    device_map="auto",
)

messages = [
    {
        "role": "user",
        "content": [
            {"type": "text", "text": "Give me a short introduction to neural networks."}
        ],
    }
]

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

with torch.no_grad():
    generated_ids = model.generate(**inputs, max_new_tokens=512)

new_tokens = generated_ids[0][inputs["input_ids"].shape[-1]:]
print(processor.decode(new_tokens, skip_special_tokens=True))

Image input example

import torch
from transformers import AutoProcessor, AutoModelForMultimodalLM

model_id = "Peymandc/Qwen3.5-0.8B-Heretic-Uncensored"

processor = AutoProcessor.from_pretrained(model_id)
model = AutoModelForMultimodalLM.from_pretrained(
    model_id,
    torch_dtype="auto",
    device_map="auto",
)

messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "Describe what you see in this image."},
        ],
    }
]

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

with torch.no_grad():
    generated_ids = model.generate(**inputs, max_new_tokens=512)

new_tokens = generated_ids[0][inputs["input_ids"].shape[-1]:]
print(processor.decode(new_tokens, skip_special_tokens=True))

Thinking mode

Qwen3.5-0.8B operates in non-thinking mode by default. Qwen's documentation says the older Qwen3 /think and /nothink soft-switch convention is not officially supported for Qwen3.5.

For an OpenAI-compatible API that supports Qwen3.5's chat-template options, thinking can be enabled with:

extra_body={"enable_thinking": True}

The exact parameter interface depends on the serving framework. Use the framework's Qwen3.5 documentation for the correct request format.

Caution: Qwen specifically notes that Qwen3.5-0.8B is more prone to entering thinking loops than larger models when thinking mode is enabled. Prefer streaming generation, set a reasonable output-token limit, and monitor generations for repetition or failure to terminate.

Sampling recommendations

The original Qwen3.5 documentation provides the following starting points. They are inherited recommendations, not parameters independently optimized for this Heretic variant.

Mode and task Temperature Top-p Top-k Presence penalty
Non-thinking, text 1.0 1.00 20 2.0
Non-thinking, vision-language 0.7 0.80 20 1.5
Thinking, text 1.0 0.95 20 1.5
Thinking, vision-language / precise coding 0.6 0.95 20 0.0

Sampling-parameter support varies between inference frameworks. Treat these as initial settings and tune them for your application. Higher presence penalties can reduce repetition but may also affect language consistency and output quality.

Other serving frameworks

The original Qwen3.5 checkpoint is documented for use with vLLM, SGLang, and KTransformers, in addition to Transformers. Compatibility of this derivative depends on the framework version and its support for the Qwen3.5 architecture. Consult the upstream Qwen3.5 model card and the relevant framework documentation for current installation and serving commands.

This repository does not separately package or claim to provide framework-specific converted checkpoints.

Performance and benchmarks

The benchmark tables published by Qwen are available in the original Qwen3.5-0.8B model card. Those numbers describe the original Qwen checkpoint, not this Heretic-modified variant.

No independent language, reasoning, vision-language, video, tool-use, or general capability benchmark results are being claimed here. The only variant-specific measurements currently reported are the Heretic refusal-keyword and KL-divergence scores in the evaluation section above.

Reproducibility

The reproduce/ directory contains the material needed to reproduce this modification. See reproduce/README.md for the exact configuration and instructions.

For a meaningful reproduction, retain the original base-model revision, Heretic version, ARA configuration, scorer configuration, evaluation datasets/splits, and relevant dependency versions. Small differences in these can change the resulting weights or reported scores.

Citation

If you use the original Qwen3.5 model, please cite the Qwen team:

@misc{qwen3.5,
  title  = {{Qwen3.5}: Towards Native Multimodal Agents},
  author = {{Qwen Team}},
  month  = {February},
  year   = {2026},
  url    = {https://qwen.ai/blog?id=qwen3.5}
}

Acknowledgements

This model is derived from Qwen/Qwen3.5-0.8B, developed by the Qwen Team.

Refusal-behavior modification was performed using Heretic v2.1.0.

Please credit both the original Qwen model and this modified release when using or redistributing this work. The upstream model's Apache 2.0 license and notices should be retained; see the original LICENSE.



💜 Support Independent Development

These models are part of my ongoing exploration of open-weight AI. Your support helps make more room for experimentation, model modifications, testing, research, and future releases.

If you find this work useful and would like to contribute, you're welcome to visit my support page:

Support my work

Every contribution is appreciated, but never expected. Sharing feedback, reporting issues, and using the models are also wonderful ways to support independent development.

Thank you for being here. ❤️

README history 9 versions

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

  1. 2026-11-20 Update README.md ae542bd 14.7 KB
    Loading...
  2. 2026-11-19 Update README.md 3c5904e 14.9 KB
    Loading...
  3. 2026-11-19 Update README.md 74c8144 14.9 KB
    Loading...
  4. 2026-11-19 Update README.md 1af2527 14.9 KB
    Loading...
  5. 2026-11-19 Update README.md 32936b0 14.7 KB
    Loading...
  6. 2026-11-19 Update README.md b8e3ddf 61 KB
    Loading...
  7. 2026-11-19 Update README.md 5daa8b1 61.1 KB
    Loading...
  8. 2026-11-19 Upload README.md with huggingface_hub 2406c0d 61.2 KB
    Loading...
  9. 2026-11-19 Upload Qwen3_5ForConditionalGeneration e98371f 5.1 KB
    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