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

cgus/granite-3.2-8b-instruct-abliterated-exl2

cgus Granite 8B second-order
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/cgus%2Fgranite-3.2-8b-instruct-abliterated-exl2"
Response includes
  • classification m1
  • files 11
  • hub_downloads_all_time 195
  • author_summary 13 models
  • readme_text full
Try next
GET/models?author=cgus
All 13 models by cgus
 GET/models?family=granite&method=m1
Sibling granite models · same method
 GET/methods/m1
What "m1" means · taxonomy
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M1
Primary method

Direct removal

No other method signals detected in this model.
Confidence
MEDIUM
Why this label 3 signals
Method inferred from partial signals - repository name, related files, or tag patterns. Producer identity not confirmed; label may sharpen or shift as we gather more evidence.
  • 'abliterated' in name/tags
  • is_gguf=0 (base model)
  • no specific method indicators - defaulting to M1 (most common)
How we label → About M1 →
Refusal direction extracted via
Extraction technique

Difference-of-means

FromArditi et al. (2024) ↗
Confidence
MEDIUM
Why we say so
primary_method=M1; difference-of-means is the reference extraction for M1/M3 (Arditi 2024)
What is a refusal direction? → Read the paper ↗
Downloads · lifetime
195
24 last 30d - stable
Likes
0
Model age
19mo ago
created 2025-03-10
Downloads over time
Now208→from3↑6,833%
0761522293 on Mar 5, 2025208 on Oct 11208 on Oct 10Mar '25Jun '25Sep '25Dec '25MarJunSep
Mar 5, 2025 → Oct 11 · 123 snapshots · spans 585 days

Genealogy 0 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.

huihui-ai/granite-3.2-8b-instruct-abliterated this lineage
cgus/granite-3.2-8b-instruct-abliterated-exl2↓ 24

Metadata

License
apache-2.0
Tags
exllamav2 granite language granite-3.2 abliterated uncensored text-generation conversational base_model:huihui-ai/granite-3.2-8b-instruct-abliterated base_model:quantized:huihui-ai/granite-3.2-8b-instruct-abliterated license:apache-2.0 4-bit

Related

Other by cgus
Qwen2-7B-Instruct-abliterated-iMat-GGUF 214 Qwen2-1.5B-Instruct-Abliterated-iMat-GGUF 195 Huihui-Qwen3-14B-abliterated-v2-exl2 57 Qwen2.5-14B-Instruct-abliterated-v2-exl2 37 Qwen2-7B-Instruct-abliterated-exl2 35
Same base model
mradermacher/granite-3.2-8b-instruct-abliterated-i1-GGUF 669 mradermacher/granite-3.2-8b-instruct-abliterated-GGUF 400 DevQuasar-5/huihui-ai.granite-3.2-8b-instruct-abliterated-GGUF 356 MaziyarPanahi/granite-3.2-8b-instruct-abliterated-GGUF 129 ikarius/Granite-3.2-8b-instruct-Abliterated-gs128-GPTQ-INT4 19
Total size
4.24 GB
Files
11
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2025-03-10 07:38
Source huggingface.co/cgus/granite-3.2-8b-instruct-abliterated-exl2 ↗ Base model huihui-ai/granite-3.2-8b-instruct-abliterated ↗

Files by quantization

Auxiliary files 11 files 4.24 GB
output.safetensors 4.24 GB db35a08b download
tokenizer.json 3.31 MB 11f3f2c6 download
vocab.json 759 KB 0a11f201 download
merges.txt 431 KB f8479fb6 download
tokenizer_config.json 8.86 KB 659def9c download
README.md 2.45 KB bf92c884 download
.gitattributes 1.48 KB a6344aac download
config.json 1.13 KB e4f2530c download
special_tokens_map.json 736 B 6811616c download
generation_config.json 144 B 83595e71 download
added_tokens.json 92.0 B 3af03016 download

README current version from Hugging Face

pipeline_tag: text-generation
inference: false
license: apache-2.0
library_name: exllamav2
tags:

  • language
  • granite-3.2
  • abliterated
  • uncensored
    base_model:
  • huihui-ai/granite-3.2-8b-instruct-abliterated

granite-3.2-8b-instruct-abliterated-exl2

Original model: granite-3.2-8b-instruct-abliterated by huihui-ai
Based on: granite-3.2-8b-instruct by Granite Team, IBM

Quants

4bpw h6 (main)
4.5bpw h6
5bpw h6
6bpw h6
8bpw h8

Quantization notes

Made with Exllamav2 0.2.8. These quants require Exllamav2 0.2.7 or newer, older versions aren't supported.
You can use these quants with TabbyAPI, Text-Generation-WebUI and some other apps.
Exl2 quants require Nvidia RTX GPU on Windows or Nvidia RTX/AMD ROCm on Linux.
Models have to fully fit your GPU VRAM because native RAM offloading isn't supported.

Original model card

huihui-ai/granite-3.2-8b-instruct-abliterated

This is an uncensored version of ibm-granite/granite-3.2-8b-instruct created with abliteration (see remove-refusals-with-transformers to know more about it).
This is a crude, proof-of-concept implementation to remove refusals from an LLM model without using TransformerLens.

Use with ollama

You can use huihui_ai/granite3.2-abliterated directly

ollama run huihui_ai/granite3.2-abliterated

Donation

If you like it, please click 'like' and follow us for more updates.
You can follow x.com/support_huihui to get the latest model information from huihui.ai.

Your donation helps us continue our further development and improvement, a cup of coffee can do it.
  • bitcoin:
  bc1qqnkhuchxw0zqjh2ku3lu4hq45hc6gy84uk70ge

README history 2 versions

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

  1. 2025-03-10Update README.mded81d1c2.5 KB
    Loading...
  2. 2025-03-10Upload README.md4909ac81.2 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.

Get the app Browse the catalog
Open in Abliteration