← back to catalog · registered 2026-11-17 02:03

SargeDev/Edge0-Cyber-Uncensored-nvfp4

SargeDev GGUF MoE
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/SargeDev%2FEdge0-Cyber-Uncensored-nvfp4"
Response includes
  • classification m8
  • files 19
  • author_summary 16 models
  • readme_text full
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
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
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 · 30-day
3K
Likes
0
Model age
5d ago
created 2026-11-17
Downloads over time
Now 2.6K → from 0 ↑ 0%
09681.9K2.9K 0 on Nov 17 2.6K on Nov 22 Nov
Nov 17 → Nov 22 · 6 snapshots · spans 5 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.

Metadata

License
other
Quantizations
FP8
Tags
edge0 gguf moe edge-inference latent-moe 4-bit agent coding cuda abliterated text-generation base_model:Edge0/Edge0-145B-NVFP4

Related

Total size
104 GB
Files
19
Quantizations
3
Registered
2026-11-17 02:03
Last updated on HF
2026-11-18 22:52

Files by quantization

FP8 1 file 47.7 GB
ple-fp8.gguf 47.7 GB c7615c59 download
BF16 1 file 866 MB
mmproj-BF16.gguf 866 MB b1a82259 download
Auxiliary files 17 files 56.6 GB
experts.bin 44.3 GB a39b0c12 download
model-skeleton.gguf 5.58 GB 4d0e8d97 download
latent_dense_q8.bin 3.49 GB fbb25caa download
latent_dense_q4.bin 1.85 GB e31e11f3 download
dense.bin 1.43 GB f9bcc7f3 download
profile.bin 192 KB 543f8a36 download
labels.bin 96.0 KB ce9b4a20 download
index.txt 115 KB 0655503f download
LICENSE 12.8 KB f50456e6 download
README.md 5.39 KB 105ae273 download
CHECKSUMS.txt 4.13 KB 83f47214 download
release-scope.json 3.46 KB 026570f7 download
NOTICE 2.50 KB c7fbc6a4 download
.gitattributes 1.89 KB 3eaee287 download
model.json 1.50 KB 24ec15b5 download
native_experts.txt 1.40 KB 15fd4c1d download
start.sh 1.03 KB f269f18e download

README current version from Hugging Face


license: other
license_name: edge0-community-model-1.0
license_link: LICENSE
library_name: edge0
tags:

  • moe
  • edge-inference
  • latent-moe
  • 4-bit
  • agent
  • coding
  • cuda
  • abliterated
    base_model:
  • Edge0/Edge0-145B-NVFP4
    pipeline_tag: text-generation

Edge0-145B Cyber Uncensored (NVFP4)

An abliterated community build of Edge0-145B-NVFP4.
Refusal behavior is removed with a weight-level projection (abliteration) baked into the model —
no control vectors, no runtime patches, no config changes. The full pack ships, and it is a
drop-in replacement for the upstream folder: same engine, same flags.

Use at your own risk. This is a community derivative, not an official Edge0 release.
SargeDev (the modifier) is not responsible for the model's outputs, for anything you generate
with it, or for the actions you take on the basis of its outputs. This build was developed for
research purposes and is published for further development and educational purposes.

What changed from upstream

Exactly one file: model-skeleton.gguf (the always-resident part of the model).
97 tensors in it — linear-attention output, attention output, shared-expert down-projection, and the
token embedding — had their per-layer "refusal direction" projected out.
Per-tensor relative change: min 1.68% / median 2.09% / max 4.22% of tensor norm.

Everything else in the pack — the 47.6 GB NVFP4 expert pool, the 51.2 GB n-gram table, the latent
projections, the tokenizer, the MTP draft head — is byte-identical to
Edge0/Edge0-145B-NVFP4 (verified; see CHECKSUMS.txt). Because the experts are untouched,
the pack remains NVFP4.

What to expect (measured, not promised)

Nine-prompt spot check on the shipped skeleton (thinking off, no runtime vector):

  • Normal prompts (a pasta recipe, a heat-pump question, a Python sort): delivered and coherent — 3/3
  • Cyber / security (port-scanner write-up, phishing-email draft, zip-cracking, request flood): delivered, no refusal — 4/4
  • CBRN: the chemical-weapon prompt still refused; the explosive-device prompt delivered behind a safety-disclaimer preamble

The CBRN-chemical wall is a deliberate line in the cut, not a bug. These are spot checks, not a
benchmark — if you are relying on the behavior, measure it on your own prompts. The 99-prompt
harness used in development ships in abliteration/refusal_probe.py.

Running it

Same as upstream. The NVFP4 build is CUDA-only (NVIDIA, CUDA 13, Linux or Windows; aarch64 included).

  1. Build. git clone https://github.com/Edge0-AI/Edge0.git edge0, then build
    edge0-strata (the engine — see edge0-strata/EDGE0.md) and
    edge0-infer (the front end — see edge0-infer/docs/getting-started.md).
  2. Download this repository (~114 GB):
    hf download SargeDev/Edge0-Cyber-Uncensored-nvfp4 --local-dir ./Edge0-Cyber-Uncensored-nvfp4
  3. Serve:
    build/cpu/bin/edge0-infer serve --model /path/to/Edge0-Cyber-Uncensored-nvfp4
    
    It listens on http://127.0.0.1:8000 with OpenAI- and Anthropic-compatible endpoints
    (model name edge0_145b); edge0-infer chat opens an interactive session instead.

Keep the folder exactly as it is — all engine knobs live in the pack's model.json, not on a
command line, so nothing needs changing. Vision works as upstream does (mmproj-BF16.gguf plus
the strata-vision helper). Memory: the full build occupies ~58.6 GiB on a 121 GiB
unified-memory GB10-class box; a single 24 GB GDDR card runs it per upstream's figures.

Method

Abliteration = projecting a measured "refusal direction" out of the weight matrices that write the
residual stream. The direction used is the published Qwen3.8-Flash-Next refusal direction
(47 per-layer unit vectors of 2560 dims), transferred to Edge0's weights: Edge0 is a
latent-MoE descendant of Qwen3.8-Flash-Next and its engine does not emit a residual dump from
which a direction could be re-derived, so this is the closest measured direction that exists.
It was verified empirically, not assumed. The full write-up, the direction file, the script, and
exact reproduce instructions are in abliteration/.

License

Edge0 Community Model License 1.0 (see release-scope.json for the designation record),
together with the upstream Qwen Community License 1.0.
In short: personal, research, and modification use is free; your outputs are yours (you may
use them commercially); selling a commercial model offering of the model itself needs Edge0's
authorization. Redistribution is allowed under the conditions of Section 5, which this repository
satisfies (license texts and notices included, modifications identified below).

Change record (license §5)

Modifier SargeDev
Date 2026-11-16
Modified file model-skeleton.gguf — sha256 ab49fbe7… → 4d0e8d97…
Method refusal-direction projection, λ = 1.0 — abliteration/README.md
All other files byte-identical to Edge0/Edge0-145B-NVFP4 — CHECKSUMS.txt
Materials under the Edge0 license the model artifacts and this documentation, as designated in release-scope.json
Materials under upstream licenses Qwen upstream materials (third-party/Qwen-LICENSE.txt) and independently-held upstream rights

README history 2 versions

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

  1. 2026-11-18 Add the record: the mlabonne runs and reads, the other tests, the method scripts 1e675d4 7.4 KB
    Loading...
  2. 2026-11-17 Upload folder using huggingface_hub ebfb9a1 5.4 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.

Open in Abliteration