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

NexesQuants/llama-3-70B-Instruct-abliterated-CQ-GGUF

NexesQuants Llama 70B GGUF 8K 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/NexesQuants%2Fllama-3-70B-Instruct-abliterated-CQ-GGUF"
Response includes
  • classification m8
  • files 6
  • hub_downloads_all_time 1,634
  • author_summary 29 models
  • readme_text full
Try next
GET/models?author=NexesQuants
All 29 models by NexesQuants
 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)

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
124 last 30d - cooling
Likes
1
Model age
2.4y ago
created 2024-05-09
Downloads over time
Now1.7K→from645↑160%
5949881.4K1.8K645 on Mar 12, 20251.7K on Oct 12Mar '25Jun '25Sep '25Dec '25MarJunSep
Mar 12, 2025 → Oct 12 · 123 snapshots · spans 579 days

Metadata

Quantizations
Q3_K
Tags
gguf endpoints_compatible region:us imatrix conversational

Related

Other by NexesQuants
Dolphin3.0-Llama3.1-1B-abliterated-GGUF 2K Llama-3.3-Nemotron-70B-Instruct-Abliterated-TA_v0.10-iMat-CQ-GGUF 96 huihui-ai_Llama-3.3-70B-Instruct-abliterated-bf16-iMat-CQ-GGUF 78 huihui-ai_Llama-3.3-70B-Instruct-abliterated-finetuned-bf16-iMat-CQ-GGUF 69 Llama_3.x_70b_L3.3-Nemotron_abliterated_fusion-iMat-CQ-GGUF 64
Total size
126 GB
Files
6
Quantizations
2
Registered
2026-08-22 13:56
Last updated on HF
2024-07-21 22:47
Source huggingface.co/NexesQuants/llama-3-70B-Instruct-abliterated-CQ-GGUF ↗

Files by quantization

Q3_K 1 file 30.9 GB
llama-3-70B-Instruct-abliterated-iMat-Q3_K_ML.gguf 30.9 GB d93f04ae download
Auxiliary files 5 files 94.6 GB
llama-3-70B-Instruct-abliterated-Imat-IQ4_MR.gguf 32.8 GB d9518a0d download
llama-3-70B-Instruct-abliterated-Imat-IQ4_SR.gguf 31.3 GB b26a873e download
llama-3-70B-Instruct-abliterated-Imat-IQ4_XSR.gguf 30.4 GB cf4fc76d download
.gitattributes 1.91 KB b66fa64e download
README.md 255 B 4416b704 download

README current version from Hugging Face

Custom GGUF quant for :
failspy/llama-3-70B-Instruct-abliterated-GGUF

IQ4_SR is Optimal (8k ctx) for 36GB VRAM with an IGP displaying the OS.

IQ4_MR is optimal for the same config with MMQ and KV quants (8 bits)

Without an IGP, the IQ4_XSR is for you.

README history 3 versions

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

  1. 2024-07-21Update README.md6f99130255 B
    Loading...
  2. 2024-05-27Update README.md60a5567183 B
    Loading...
  3. 2024-05-09Create README.md762a321132 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