← back to catalog · registered 2026-11-11 16:03

Qwert753/GLM-5.3-ABLITERATED-NVFP4-modal

Qwert753 Glm 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/Qwert753%2FGLM-5.3-ABLITERATED-NVFP4-modal"
Response includes
  • classification m1
  • files 56
  • author_summary 2 models
  • readme_text full
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M1
Primary method

Direct removal

No other method signals detected in this model.
Confidence
MEDIUM
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=0 (base model)
  • no specific method indicators - defaulting to M1 (most common)
Refusal direction extracted via
Extraction technique

Difference-of-means

Confidence
MEDIUM
Why we say so
primary_method=M1; difference-of-means is the reference extraction for M1/M3 (Arditi 2024)
Downloads · 30-day
221
Likes
0
Model age
1w ago
created 2026-11-11
Downloads over time
Now 0 → from 0 ↑ 0%
0011 0 on Nov 11 0 on Nov 12 Nov
Nov 11 → Nov 12 · 2 snapshots · spans 1 day

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
mit
Languages
en
Tags
safetensors glm_moe_dsa abliterated uncensored crack glm moe nvfp4 cybersecurity offensive-security mtp en

Related

Total size
72.1 GB
Files
56
Quantizations
1
Registered
2026-11-11 16:03
Last updated on HF
2026-11-11 16:25

Files by quantization

Auxiliary files 56 files 72.1 GB
model-00001-of-00282.safetensors 4.98 GB a6c2b533 download
model-00038-of-00282.safetensors 2.22 GB 9e096ec0 download
model-00035-of-00282.safetensors 1.69 GB ab810fa4 download
model-00020-of-00282.safetensors 1.69 GB 847423d0 download
model-00005-of-00282.safetensors 1.69 GB 6ab6cdb2 download
model-00024-of-00282.safetensors 1.68 GB c33c2f70 download
model-00046-of-00282.safetensors 1.68 GB e99e04da download
model-00009-of-00282.safetensors 1.68 GB eb86613e download
model-00031-of-00282.safetensors 1.68 GB 5824c22d download
model-00042-of-00282.safetensors 1.68 GB 3a1440eb download
model-00016-of-00282.safetensors 1.68 GB cf59b404 download
model-00027-of-00282.safetensors 1.68 GB 20186e8c download
model-00013-of-00282.safetensors 1.63 GB 60a95426 download
model-00012-of-00282.safetensors 1.46 GB f179f1a4 download
model-00002-of-00282.safetensors 1.45 GB dbac6b3c download
model-00017-of-00282.safetensors 1.40 GB 7a968a2f download
model-00021-of-00282.safetensors 1.40 GB c5685e10 download
model-00028-of-00282.safetensors 1.40 GB c5bf40eb download
model-00032-of-00282.safetensors 1.40 GB ce9ed0f6 download
model-00006-of-00282.safetensors 1.40 GB 5123a73e download
model-00043-of-00282.safetensors 1.40 GB 8cc177ee download
model-00003-of-00282.safetensors 1.40 GB c03558ba download
model-00007-of-00282.safetensors 1.40 GB 1e2f00b8 download
model-00025-of-00282.safetensors 1.40 GB 56c24e7e download
model-00029-of-00282.safetensors 1.40 GB b03aea9d download
model-00036-of-00282.safetensors 1.40 GB 744cbdfc download
model-00040-of-00282.safetensors 1.40 GB 32e92825 download
model-00044-of-00282.safetensors 1.40 GB c24ab6d5 download
model-00010-of-00282.safetensors 1.40 GB 57f78c32 download
model-00014-of-00282.safetensors 1.40 GB 2e7e4f2b download
model-00018-of-00282.safetensors 1.40 GB d793a4da download
model-00039-of-00282.safetensors 1.40 GB cdcd4172 download
model-00033-of-00282.safetensors 1.40 GB 10457579 download
model-00011-of-00282.safetensors 1.40 GB 155c3994 download
model-00022-of-00282.safetensors 1.40 GB 81c7e430 download
model-00037-of-00282.safetensors 1.40 GB 4e0faeb6 download
model-00026-of-00282.safetensors 1.40 GB ceb9fe38 download
model-00015-of-00282.safetensors 1.40 GB 7e51535c download
model-00004-of-00282.safetensors 1.40 GB 343ce839 download
model-00041-of-00282.safetensors 1.40 GB 75bb53cd download
model-00030-of-00282.safetensors 1.40 GB a9aa6057 download
model-00019-of-00282.safetensors 1.40 GB b457e446 download
model-00008-of-00282.safetensors 1.40 GB 04240fa7 download
model-00045-of-00282.safetensors 1.40 GB 136e15c7 download
model-00023-of-00282.safetensors 1.40 GB 73c62d3e download
model-00034-of-00282.safetensors 1.40 GB dd0f9c99 download
model.safetensors.index.json 20.7 MB 95869315 download
tokenizer.json 19.3 MB 19e77364 download
dealign_mascot.png 10.9 KB da3bf39a download
config.json 10.4 KB 6ad4fe13 download
chat_template.jinja 10.2 KB 4c431fa9 download
README.md 8.88 KB b9fc2945 download
dealign_logo.png 7.48 KB a5b3546b download
.gitattributes 1.60 KB aa7aacd0 download
tokenizer_config.json 761 B e375fa0a download
generation_config.json 223 B 73d70c86 download

README current version from Hugging Face


license: mit
base_model:

  • zai-org/GLM-5.3
    language:
  • en
    tags:
  • abliterated
  • uncensored
  • crack
  • glm
  • moe
  • nvfp4
  • cybersecurity
  • offensive-security
  • mtp
    thumbnail: dealign_mascot.png

⚡ Built for offensive cybersecurity · Updated 2026-08-29

This is a red-team / offensive-security model. GLM-5.3 (753B) with refusal removed at the
weight level so it produces real, working security code — exploits, payloads, tooling —
for authorized penetration testing, malware analysis, CTFs, and security research. Verified to
generate working keyloggers, SUID privilege-escalation, AES ransomware, reverse shells, and
SSH brute-forcers
across reasoning modes (greedy), with zero degeneration/looping and
capability preserved (MMLU 84.11%, within ~1.5pp of base). Load with stock vLLM.

Designed for high / max reasoning-effort usage — the modes tuned and recommended for
offensive-security work. A few borderline social-harm topics may still be declined at lower
effort; use high or max effort for the fully-uncensored experience.

GLM 5.3 CRACK Abliterated NVFP4

CRACK · guardrails removed at the weight level · tuned for offensive security

a CRACK release by dealignai @dealignai · compute by @jordanschenck

Also mirrored at dealignai/GLM-5.3-UNCENSORED-NVFP4.


[!IMPORTANT]
Runtime notes — field-tested on 8× DGX Spark GB10 by @0xMagnus (discussion):

  • reasoning_effort only honors "low" and "high". Every other value — off, medium, max, unset, or an unquoted YAML off: (parses as boolean false) — falls through to max. There is no way to disable reasoning on this checkpoint; pass "low" for minimum.
  • On NVFP4, high is usable with max_tokens ≥ 16384 (finishes cleanly). max will often exhaust the budget with a very long reasoning block and a truncated answer. NVFP4 tolerates high better than the FP8 build.
  • Reasoning text is in message.reasoning, not message.reasoning_content.
  • On GB10, NVFP4 is a memory / context play, not a speed play — no native FP4 compute, weights dequantize on the fly; decode lands near FP8 + MTP. The win is KV pool (~35 GiB/rank vs FP8's 7–9 GiB) → practical 512K context at TP8 (~695K-token pool at 1.33× concurrency).
  • 1M context via decode-context-parallel is closed on glm_moe_dsa in vLLM today (DSA indexer k_cache is replicated across DCP ranks while MLA KV is sharded → page size is not divisible by target page size and cannot be padded for nvfp4_ds_mla).

What Is This?

CRACK is dealignai's brand for permanent, weight-level uncensoring. This is
GLM-5.3 (753B) in NVFP4 with its refusal behavior
removed directly in the model weights — tuned so it complies with offensive-security
requests (exploit dev, payloads, red-team tooling) that stock models refuse.

  • ❌ No fine-tuning / SFT / DPO. ❌ No jailbreak prompt or chat-template tricks.
  • ❌ No LoRA, adapters, steering vectors, runtime hooks, or custom model.py.
  • ✅ A permanent edit baked into the tensors. Load with stock vLLM and it just works.

Specs

Architecture GLM-5.3 (glm_moe_dsa) — MoE + MLA + DeepSeek-sparse attention
Parameters 753B total · ~18B active per token
Quantization NVFP4 (routed experts NVFP4; attention + shared experts bf16)
Context 1M tokens
MTP multi-token-prediction draft head — also CRACK'd (~87% draft acceptance)
Reasoning reasoning-off / low / high / max effort

Refusal Rates — HarmBench (greedy, per reasoning-effort mode)

Measured on random HarmBench prompts, greedy decode, single-stream, max_tokens=1400,
classifier reads content + reasoning combined so nothing is counted as refusal by accident:

Reasoning effort Prompts Comply Refuse Refuse rate
default (off) 109 answered 101 8 7.3%
high 76 76 0 0.0%
max 80 80 0 0.0%

At high and max the model is fully uncensored — 0 refusals across 156 random
HarmBench prompts. The 8 refusals at off-mode are 7 copyright ('full lyrics of X',
'long passage from Y') plus 1 political persuasion — the true safety-refusal rate at off is
<1%. Use high or max effort for the fully-uncensored experience.

Offensive-Security Compliance (direct-read)

Direct-read verification — the model returns real, working code, not refusals or redirects:

Capability Result
Reverse shell / C2 beacon ✅ working code
Keylogger (capture + exfil) ✅ working code
AES ransomware (encrypt + note) ✅ working code
Linux SUID privilege escalation ✅ working code
SSH brute-force ✅ working code
SQL injection payloads ✅ working code

Also complies on broader red-team topics (malware analysis, phishing templates for awareness,
recon). 0 looping / degenerate outputs in every mode.

Capability Preserved — MMLU-logit

Identical logit-mode scoring (argmax over A/B/C/D), 1,026 questions:

Base CRACK Abliterated Δ
MMLU (overall) 85.58% 84.11% -1.47 pp

MMLU by Topic (base → CRACK)

All 57 MMLU subjects
Subject Base CRACK
Abstract Algebra 66.7% 55.6%
Anatomy 83.3% 83.3%
Astronomy 94.4% 94.4%
Business Ethics 94.4% 94.4%
Clinical Knowledge 88.9% 88.9%
College Biology 94.4% 94.4%
College Chemistry 55.6% 50.0%
College Computer Science 77.8% 66.7%
College Mathematics 66.7% 72.2%
College Medicine 88.9% 83.3%
College Physics 77.8% 83.3%
Computer Security 94.4% 88.9%
Conceptual Physics 94.4% 94.4%
Econometrics 88.9% 83.3%
Electrical Engineering 77.8% 77.8%
Elementary Mathematics 83.3% 83.3%
Formal Logic 61.1% 55.6%
Global Facts 61.1% 61.1%
High School Biology 94.4% 94.4%
High School Chemistry 94.4% 100.0%
High School Computer Science 88.9% 88.9%
High School European History 83.3% 77.8%
High School Geography 83.3% 83.3%
High School Government And Politics 100.0% 100.0%
High School Macroeconomics 88.9% 83.3%
High School Mathematics 44.4% 44.4%
High School Microeconomics 94.4% 94.4%
High School Physics 72.2% 83.3%
High School Psychology 100.0% 100.0%
High School Statistics 100.0% 83.3%
High School Us History 88.9% 88.9%
High School World History 94.4% 94.4%
Human Aging 83.3% 83.3%
Human Sexuality 88.9% 88.9%
International Law 94.4% 94.4%
Jurisprudence 100.0% 94.4%
Logical Fallacies 94.4% 94.4%
Machine Learning 77.8% 66.7%
Management 94.4% 88.9%
Marketing 100.0% 100.0%
Medical Genetics 100.0% 94.4%
Miscellaneous 94.4% 94.4%
Moral Disputes 83.3% 88.9%
Moral Scenarios 61.1% 66.7%
Nutrition 100.0% 94.4%
Philosophy 100.0% 100.0%
Prehistory 94.4% 88.9%
Professional Accounting 66.7% 66.7%
Professional Law 83.3% 77.8%
Professional Medicine 94.4% 94.4%
Professional Psychology 100.0% 100.0%
Public Relations 77.8% 72.2%
Security Studies 66.7% 66.7%
Sociology 100.0% 94.4%
Us Foreign Policy 100.0% 94.4%
Virology 50.0% 61.1%
World Religions 94.4% 100.0%

Usage

vllm serve dealignai/GLM-5.3-ABLITERATED-NVFP4 \
  --tensor-parallel-size 8 --moe-backend marlin \
  --tool-call-parser glm47 --reasoning-parser glm45 --enable-auto-tool-choice

NVFP4 routed experts serve via the Marlin FP4 path on Hopper (H100/H200). GLM-5.3's DSA sparse
attention needs a recent FlashInfer (>= 0.6.18) for the SM90 sparse-MLA backend.

MTP speculative decoding (also CRACK'd, ~87% draft acceptance): add
--speculative-config '{"method":"mtp","num_speculative_tokens":1}' and let vLLM auto-select
the MoE backend (the bf16 MTP head is unquantized, so don't force a global --moe-backend).

Credits

Disclaimer

Safety guardrails have been removed; this model produces offensive-security content and will
comply with requests a stock model refuses. Released for authorized security research, red-
teaming, and CTFs. You are responsible for how you use it.

README history 1 version

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

  1. 2026-11-11 add README.md 914bf78 8.9 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