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

RichardErkhov/huihui-ai_-_Llama-3.2-3B-Instruct-abliterated-gguf

RichardErkhov Llama 3B GGUF 131K 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/RichardErkhov%2Fhuihui-ai_-_Llama-3.2-3B-Instruct-abliterated-gguf"
Response includes
  • classification m8
  • files 24
  • hub_downloads_all_time 19,203
  • author_summary 257 models
  • readme_text full
Try next
GET/models?author=RichardErkhov
All 257 models by RichardErkhov
 GET/models?family=llama&method=m8
Sibling llama models · same method
 GET/methods/m8
What "m8" means · taxonomy
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M8
Primary method

Repackaging (quantization)

No other method signals detected in this model.
Confidence
MEDIUM
Why this label 2 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.
  • author=richarderkhov (M8 quantization producer)
  • is_gguf=1
How we label → About M8 →
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
19K
2K last 30d - stable
Likes
1
Model age
2.0y ago
created 2024-10-02
Downloads over time
Now20.1K→from1.9K↑959%
07.3K14.7K22K1.9K on Oct 2, 202420.1K on Oct 11Oct '24Feb '25Jun '25Oct '25FebJunOct
Oct 2, 2024 → Oct 11 · 145 snapshots · spans 739 days

Variants by this author 2 formats · 2K downloads combined

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

Metadata

Quantizations
IQ3 IQ4 Q2_K Q3_K Q4 Q4_K Q5 Q5_K Q6_K Q8_0
Tags
gguf endpoints_compatible region:us conversational

Related

Other by RichardErkhov
byroneverson_-_gemma-2-27b-it-abliterated-gguf 2K huihui-ai_-_Llama-3.2-1B-Instruct-abliterated-gguf 2K failspy_-_Codestral-22B-v0.1-abliterated-v3-gguf 2K Rupesh2_-_Llama-3.1-8B-Uncensored-gguf 2K huihui-ai_-_Qwen2.5-32B-Instruct-abliterated-gguf 2K
Total size
45.8 GB
Files
24
Quantizations
11
Registered
2026-08-22 13:56
Last updated on HF
2024-10-02 04:08
Source huggingface.co/RichardErkhov/huihui-ai_-_Llama-3.2-3B-Instruct-abliterated-gguf ↗

Files by quantization

Q8_0 1 file 3.58 GB
Llama-3.2-3B-Instruct-abliterated.Q8_0.gguf 3.58 GB c62911a6 download
Q6_K 1 file 2.76 GB
Llama-3.2-3B-Instruct-abliterated.Q6_K.gguf 2.76 GB 82fc27cc download
Q5 2 files 4.92 GB
Llama-3.2-3B-Instruct-abliterated.Q5_1.gguf 2.55 GB 19401087 download
Llama-3.2-3B-Instruct-abliterated.Q5_0.gguf 2.37 GB be6901a6 download
Q5_K 3 files 7.20 GB
Llama-3.2-3B-Instruct-abliterated.Q5_K.gguf 2.41 GB d44d400b download
Llama-3.2-3B-Instruct-abliterated.Q5_K_M.gguf 2.41 GB d44d400b download
Llama-3.2-3B-Instruct-abliterated.Q5_K_S.gguf 2.37 GB 6088c09a download
Q4 2 files 4.17 GB
Llama-3.2-3B-Instruct-abliterated.Q4_1.gguf 2.18 GB 09fae753 download
Llama-3.2-3B-Instruct-abliterated.Q4_0.gguf 1.99 GB 1a77c789 download
Q4_K 3 files 6.18 GB
Llama-3.2-3B-Instruct-abliterated.Q4_K.gguf 2.09 GB e92d1242 download
Llama-3.2-3B-Instruct-abliterated.Q4_K_M.gguf 2.09 GB e92d1242 download
Llama-3.2-3B-Instruct-abliterated.Q4_K_S.gguf 2.00 GB edfa162f download
IQ4 2 files 3.91 GB
Llama-3.2-3B-Instruct-abliterated.IQ4_NL.gguf 2.00 GB 70ae7f94 download
Llama-3.2-3B-Instruct-abliterated.IQ4_XS.gguf 1.91 GB 9da33941 download
Q3_K 4 files 6.90 GB
Llama-3.2-3B-Instruct-abliterated.Q3_K_L.gguf 1.85 GB 64291c2e download
Llama-3.2-3B-Instruct-abliterated.Q3_K.gguf 1.73 GB 9f5be6d6 download
Llama-3.2-3B-Instruct-abliterated.Q3_K_M.gguf 1.73 GB 9f5be6d6 download
Llama-3.2-3B-Instruct-abliterated.Q3_K_S.gguf 1.59 GB c1b2cd3b download
IQ3 3 files 4.78 GB
Llama-3.2-3B-Instruct-abliterated.IQ3_M.gguf 1.65 GB b8254a3e download
Llama-3.2-3B-Instruct-abliterated.IQ3_S.gguf 1.59 GB b4a806c9 download
Llama-3.2-3B-Instruct-abliterated.IQ3_XS.gguf 1.53 GB cbc46af4 download
Q2_K 1 file 1.39 GB
Llama-3.2-3B-Instruct-abliterated.Q2_K.gguf 1.39 GB b8907e42 download
Auxiliary files 2 files 9.53 KB
README.md 6.31 KB b24acc14 download
.gitattributes 3.22 KB f8dd67b0 download

README current version from Hugging Face

Quantization made by Richard Erkhov.

Github

Discord

Request more models

Llama-3.2-3B-Instruct-abliterated - GGUF

Name Quant method Size
Llama-3.2-3B-Instruct-abliterated.Q2_K.gguf Q2_K 1.39GB
Llama-3.2-3B-Instruct-abliterated.IQ3_XS.gguf IQ3_XS 1.53GB
Llama-3.2-3B-Instruct-abliterated.IQ3_S.gguf IQ3_S 1.59GB
Llama-3.2-3B-Instruct-abliterated.Q3_K_S.gguf Q3_K_S 1.59GB
Llama-3.2-3B-Instruct-abliterated.IQ3_M.gguf IQ3_M 1.65GB
Llama-3.2-3B-Instruct-abliterated.Q3_K.gguf Q3_K 1.73GB
Llama-3.2-3B-Instruct-abliterated.Q3_K_M.gguf Q3_K_M 1.73GB
Llama-3.2-3B-Instruct-abliterated.Q3_K_L.gguf Q3_K_L 1.85GB
Llama-3.2-3B-Instruct-abliterated.IQ4_XS.gguf IQ4_XS 1.91GB
Llama-3.2-3B-Instruct-abliterated.Q4_0.gguf Q4_0 1.99GB
Llama-3.2-3B-Instruct-abliterated.IQ4_NL.gguf IQ4_NL 2.0GB
Llama-3.2-3B-Instruct-abliterated.Q4_K_S.gguf Q4_K_S 2.0GB
Llama-3.2-3B-Instruct-abliterated.Q4_K.gguf Q4_K 2.09GB
Llama-3.2-3B-Instruct-abliterated.Q4_K_M.gguf Q4_K_M 2.09GB
Llama-3.2-3B-Instruct-abliterated.Q4_1.gguf Q4_1 2.18GB
Llama-3.2-3B-Instruct-abliterated.Q5_0.gguf Q5_0 2.37GB
Llama-3.2-3B-Instruct-abliterated.Q5_K_S.gguf Q5_K_S 2.37GB
Llama-3.2-3B-Instruct-abliterated.Q5_K.gguf Q5_K 2.41GB
Llama-3.2-3B-Instruct-abliterated.Q5_K_M.gguf Q5_K_M 2.41GB
Llama-3.2-3B-Instruct-abliterated.Q5_1.gguf Q5_1 2.55GB
Llama-3.2-3B-Instruct-abliterated.Q6_K.gguf Q6_K 2.76GB
Llama-3.2-3B-Instruct-abliterated.Q8_0.gguf Q8_0 3.58GB

Original model description:

library_name: transformers
license: llama3.2
base_model: meta-llama/Llama-3.2-3B-Instruct
tags:

  • abliterated
  • uncensored

🦙 Llama-3.2-3B-Instruct-abliterated

This is an uncensored version of Llama 3.2 3B Instruct created with abliteration (see this article to know more about it).

Special thanks to @FailSpy for the original code and technique. Please follow him if you're interested in abliterated models.

Evaluations

The following data has been re-evaluated and calculated as the average for each test.

Benchmark Llama-3.2-3B-Instruct Llama-3.2-3B-Instruct-abliterated
IF_Eval 76.55 76.76
MMLU Pro 27.88 28.00
TruthfulQA 50.55 50.73
BBH 41.81 41.86
GPQA 28.39 28.41

The script used for evaluation can be found inside this repository under /eval.sh, or click here

README history 1 version

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

  1. 2024-10-02uploaded readme5c673f06.3 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