Get the app
← back to catalog · registered 2026-08-22 13:56

Exil01/Qwen3.5-27B-Uncensored-YaRN-1M

Exil01 Qwen 27B GGUF 262K ctx
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/Exil01%2FQwen3.5-27B-Uncensored-YaRN-1M"
Response includes
  • classification m-uncensored
  • files 6
  • hub_downloads_all_time 2,755
  • author_summary 7 models
  • readme_text full
Try next
GET/models?author=Exil01
All 7 models by Exil01
 GET/models?family=qwen&method=m-uncensored
Sibling qwen models · same method
 GET/methods/m-uncensored
What "m-uncensored" means · taxonomy
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M-U
Primary method

Uncensored (method unknown)

No other method signals detected in this model.
Confidence
LOW
Why this label 3 signals
Weak or ambiguous signals. Best guess based on catalog patterns; treat as tentative and check the evidence below.
  • 'uncensored' in name/tags but no 'abliterated' marker
  • method not identifiable from author declaration alone
  • may be DPO fine-tune, prompt engineering, or unknown technique
How we label → About M-U →
Refusal direction extraction

No specific extraction method could be identified for this model. The producer either did not document it or used a proprietary pipeline.

What is a refusal direction? →
Downloads · lifetime
3K
113 last 30d - cooling
Likes
2
Model age
6mo ago
created 2026-03-16
Downloads over time
Now2.8K→from828↑236%
7301.5K2.2K3K828 on Mar 182.8K on Oct 11MarAprMayJunJulAugSepOct
Mar 18 → Oct 11 · 69 snapshots · spans 207 days

Metadata

License
apache-2.0
Quantizations
BF16
Tags
gguf qwen3_5 yarn long-context uncensored text-generation en arxiv:2309.00071 license:apache-2.0 endpoints_compatible region:us conversational

Related

Other by Exil01
Qwen3.5-35B-Uncensored-YaRN-1M 635 Qwen3.5-35B-Uncensored 477 Huihui-Qwen3.5-27B-Opus-abliterated-YaRN-1M 209 Josiefied-Qwen3-0.6B-abliterated-v1-Q4_K_M-GGUF 38 Josiefied-Qwen3-0.6B-abliterated-v1-Q8_0-GGUF 11
Total size
50.1 GB
Files
6
Quantizations
3
Registered
2026-08-22 13:56
Last updated on HF
2026-03-16 21:07
Source huggingface.co/Exil01/Qwen3.5-27B-Uncensored-YaRN-1M ↗

Files by quantization

BF16 1 file 50.1 GB
Qwen3.5-27B-Uncensored-HauhauCS-Aggressive-BF16.gguf 50.1 GB 4987cf92 download
F16 1 file 885 MB
mmproj-Qwen3.5-27B-Uncensored-HauhauCS-Aggressive-f16.gguf 885 MB fcd4e139 download
Auxiliary files 4 files 12.2 MB
tokenizer.json 12.2 MB 5f9e4d49 download
config.json 4.04 KB bde2be7f download
.gitattributes 1.71 KB 44d1f6d3 download
README.md 927 B 5c2054ed download

README current version from Hugging Face

license: apache-2.0
language: en
pipeline_tag: text-generation
tags: [yarn, long-context, uncensored]

Qwen3.5-27B-Uncensored-YaRN-1M

Uncensored Qwen3.5 27B with aggressive HauhauCS modifications and YaRN extension, supporting up to 1M context.

Model Details

  • Base: HauhauCS/Qwen3.5-27B-Uncensored-HauhauCS-Aggressive
  • Context: 262K active / 1M max (YaRN)
  • Quantization: BF16
  • VRAM: ~60GB (262K) / ~90GB (1M)

Usage

262K Context (80GB GPU)

1M Context (100GB+ GPU or Q8_0 model)

With Vision (mmproj)

Files

  • GGUF: 53.8GB (BF16)
  • mmproj: 885MB (F16, vision)
  • Tokenizer: 13MB

Performance

  • 262K context: ~60GB VRAM (BF16 cache)
  • 1M context: ~90GB VRAM (BF16 cache)
  • Max output: 81,920 tokens
  • Temperature: 0.7 (uncensored optimized)

Credits

  • Base: HauhauCS (uncensored aggressive)
  • Original: Qwen Team
  • YaRN: arxiv.org/abs/2309.00071

README history 2 versions

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

  1. 2026-03-16Upload README.md with huggingface_hube2ea3a2927 B
    Loading...
  2. 2026-03-16Upload README.md with huggingface_huba2adfff632 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.

Get the app Browse the catalog
Open in Abliteration