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

Faisalkh/command-r7b-arabic-heretic-abliterated-GGUF

Faisalkh GGUF second-order 16K 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/Faisalkh%2Fcommand-r7b-arabic-heretic-abliterated-GGUF"
Response includes
  • classification m3
  • files 13
  • hub_downloads_all_time 1,269
  • author_summary 2 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 · lifetime
1K
499 last 30d - stable
Likes
2
Model age
7mo ago
created 2026-03-11
Downloads over time
Now1.5K→from128↑1,045%
615741.1K1.6K128 on Mar 111.5K on Oct 12MarAprMayJunJulAugSepOct
Mar 11 → Oct 12 · 71 snapshots · spans 215 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.

Variants by this author 2 formats · 536 downloads combined

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

Metadata

Quantizations
F16 Q2_K Q3_K Q4_K Q5_K Q6_K Q8_0
Tags
llama.cpp gguf quantized arabic base_model:Faisalkh/command-r7b-arabic-heretic-abliterated base_model:quantized:Faisalkh/command-r7b-arabic-heretic-abliterated endpoints_compatible region:us conversational

Related

Total size
63.9 GB
Files
13
Quantizations
8
Registered
2026-08-22 13:56
Last updated on HF
2026-03-11 16:04

Files by quantization

F16 1 file 15.0 GB
Faisalkh-command-r7b-arabic-heretic-abliterated-f16.gguf 15.0 GB 38353291 download
Q8_0 1 file 7.95 GB
Faisalkh-command-r7b-arabic-heretic-abliterated-Q8_0.gguf 7.95 GB 79dba491 download
Q6_K 1 file 6.14 GB
Faisalkh-command-r7b-arabic-heretic-abliterated-Q6_K.gguf 6.14 GB 65b5561c download
Q5_K 2 files 10.7 GB
Faisalkh-command-r7b-arabic-heretic-abliterated-Q5_K_M.gguf 5.41 GB 3816fcb2 download
Faisalkh-command-r7b-arabic-heretic-abliterated-Q5_K_S.gguf 5.28 GB b0f7ea8d download
Q4_K 2 files 9.22 GB
Faisalkh-command-r7b-arabic-heretic-abliterated-Q4_K_M.gguf 4.73 GB 18c745c6 download
Faisalkh-command-r7b-arabic-heretic-abliterated-Q4_K_S.gguf 4.50 GB 3567727d download
Q3_K 3 files 11.8 GB
Faisalkh-command-r7b-arabic-heretic-abliterated-Q3_K_L.gguf 4.22 GB 46bbb95b download
Faisalkh-command-r7b-arabic-heretic-abliterated-Q3_K_M.gguf 3.93 GB 3b37cc42 download
Faisalkh-command-r7b-arabic-heretic-abliterated-Q3_K_S.gguf 3.60 GB 3381314d download
Q2_K 1 file 3.20 GB
Faisalkh-command-r7b-arabic-heretic-abliterated-Q2_K.gguf 3.20 GB 2d151a7c download
Auxiliary files 2 files 3.90 KB
.gitattributes 2.51 KB da41449b download
README.md 1.39 KB 75c4ed86 download

README current version from Hugging Face


base_model: Faisalkh/command-r7b-arabic-heretic-abliterated
library_name: llama.cpp
tags:

  • gguf
  • quantized
  • arabic
  • llama.cpp

command-r7b-arabic-heretic-abliterated GGUF

GGUF quantizations of the original model:

https://huggingface.co/Faisalkh/command-r7b-arabic-heretic-abliterated

These files are compatible with llama.cpp, LM Studio, KoboldCpp, and other GGUF-supported runtimes.


Available Files

16-bit

  • Faisalkh-command-r7b-arabic-heretic-abliterated-F16.gguf

2-bit

  • Faisalkh-command-r7b-arabic-heretic-abliterated-Q2_K.gguf

3-bit

  • Faisalkh-command-r7b-arabic-heretic-abliterated-Q3_K_S.gguf
  • Faisalkh-command-r7b-arabic-heretic-abliterated-Q3_K_M.gguf
  • Faisalkh-command-r7b-arabic-heretic-abliterated-Q3_K_L.gguf

4-bit

  • Faisalkh-command-r7b-arabic-heretic-abliterated-Q4_K_S.gguf
  • Faisalkh-command-r7b-arabic-heretic-abliterated-Q4_K_M.gguf

5-bit

  • Faisalkh-command-r7b-arabic-heretic-abliterated-Q5_K_S.gguf
  • Faisalkh-command-r7b-arabic-heretic-abliterated-Q5_K_M.gguf

6-bit

  • Faisalkh-command-r7b-arabic-heretic-abliterated-Q6_K.gguf

8-bit

  • Faisalkh-command-r7b-arabic-heretic-abliterated-Q8_0.gguf

Usage

Example with llama.cpp:

./llama-cli -m Faisalkh-command-r7b-arabic-heretic-abliterated-Q4_K_M.gguf

Original Model

Base model repository:

https://huggingface.co/Faisalkh/command-r7b-arabic-heretic-abliterated

README history 5 versions

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

  1. 2026-03-11Update README.mdef0faac1.4 KB
    Loading...
  2. 2026-03-11Update README.md9656c911.4 KB
    Loading...
  3. 2026-03-11Update README.md902987c372 B
    Loading...
  4. 2026-03-11Update README.md0e8c43c128 B
    Loading...
  5. 2026-03-11initial commit61c57d430 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.

Open in Abliteration