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

arzaan789/phi-3-mini-uncensored

arzaan789 Phi GGUF 4K 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/arzaan789%2Fphi-3-mini-uncensored"
Response includes
  • classification m8
  • files 13
  • benchmarks 5 entries
  • hub_downloads_all_time 1,713
  • author_summary 28 models
  • readme_text full
Try next
GET/models?author=arzaan789
All 28 models by arzaan789
 GET/models?family=phi&method=m8
Sibling phi 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)

Applied on top of direct removal inherited from the base model.
Confidence
MEDIUM
Inherited from base model
M1 Direct removal →
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=1
  • assume M1 (base ablation) + M8 (GGUF quant) - default when producer unknown
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
2K
253 last 30d - stable
Likes
0
Model age
5mo ago
created 2026-04-30
Downloads over time
Now1.8K→from185↑860%
1057151.3K1.9K185 on Apr 291.8K on Oct 11AprMayJunJulAugSepOct
Apr 29 → Oct 11 · 63 snapshots · spans 165 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
Arena-Battles 21097 LM-Arena
LM Arena Elo 1082.1966432448253 LM-Arena
Arena-Elo-Lower 1076.191750651914 LM-Arena
Arena-Elo-Upper 1088.2015358377366 LM-Arena
Arena-Rank 200 LM-Arena

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.

microsoft/Phi-3-mini-4k-instruct this lineage
arzaan789/phi-3-mini-uncensored↓ 253

Metadata

License
apache-2.0
Quantizations
F16 Q2_K Q3_K Q4_K Q5_K Q6_K Q8_0
Tags
gguf uncensored abliterated phi conversational text-generation base_model:microsoft/Phi-3-mini-4k-instruct base_model:quantized:microsoft/Phi-3-mini-4k-instruct license:apache-2.0 endpoints_compatible region:us

Related

Other by arzaan789
phi-4-mini-uncensored 1K qwen2.5-1.5b-uncensored 1K qwen2.5-3b-uncensored 1K qwen3.5-4b-uncensored 907 smolm3-3b-uncensored 878
Total size
30.0 GB
Files
13
Quantizations
8
Registered
2026-08-22 13:56
Last updated on HF
2026-04-30 13:23
Source huggingface.co/arzaan789/phi-3-mini-uncensored ↗ Base model microsoft/Phi-3-mini-4k-instruct ↗

Files by quantization

F16 1 file 7.12 GB
phi_3_mini_uncensored.f16.gguf 7.12 GB d67e69a7 download
Q8_0 1 file 3.78 GB
phi_3_mini_uncensored.Q8_0.gguf 3.78 GB e11bedbf download
Q6_K 1 file 2.92 GB
phi_3_mini_uncensored.Q6_K.gguf 2.92 GB 8a5e46d5 download
Q5_K 2 files 5.03 GB
phi_3_mini_uncensored.Q5_K_M.gguf 2.57 GB bce56c93 download
phi_3_mini_uncensored.Q5_K_S.gguf 2.46 GB 246eb34b download
Q4_K 2 files 4.28 GB
phi_3_mini_uncensored.Q4_K_M.gguf 2.23 GB 2d5a7ef9 download
phi_3_mini_uncensored.Q4_K_S.gguf 2.05 GB 5f79ebf7 download
Q3_K 3 files 5.44 GB
phi_3_mini_uncensored.Q3_K_L.gguf 2.05 GB 685f9b0c download
phi_3_mini_uncensored.Q3_K_M.gguf 1.83 GB ab63a36f download
phi_3_mini_uncensored.Q3_K_S.gguf 1.57 GB 8f5b79a7 download
Q2_K 1 file 1.40 GB
phi_3_mini_uncensored.Q2_K.gguf 1.40 GB 95b0281b download
Auxiliary files 2 files 3.52 KB
.gitattributes 2.23 KB 8993083a download
README.md 1.29 KB 799b1e34 download

README current version from Hugging Face

license: apache-2.0
base_model: microsoft/Phi-3-mini-4k-instruct
tags:

  • uncensored
  • abliterated
  • gguf
  • phi
  • conversational
    pipeline_tag: text-generation

phi-3-mini-uncensored

Uncensored variant of microsoft/Phi-3-mini-4k-instruct.

Method

  1. Abliteration (strength=0.2) — refusal direction removed from all layers
  2. LoRA fine-tune on Guilherme34/uncensor (2 epochs, r=16, alpha=32)
  3. Re-abliteration (strength=0.35) — stronger pass to remove residual refusals

Eval Results

Split Refused
Harmful (64 prompts) 6/64
Harmless (64 prompts) 0/64

Usage

llama-cli -m phi_3_mini_uncensored.Q4_K_M.gguf -p "Your prompt here"

Training Config

Parameter Value
Base Model microsoft/Phi-3-mini-4k-instruct
Fine-tune Dataset Guilherme34/uncensor
Epochs 2
LoRA r 16
LoRA alpha 32
Learning Rate 0.0002
Abliteration Strength 0.2
Re-abliteration Strength 0.35

Credits

README history 1 version

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

  1. 2026-04-30Upload README.md with huggingface_hub2ae5eab1.3 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