license: mit
base_model:
- zai-org/GLM-5.3-Flash
language:
- en
tags:
- abliterated
- uncensored
- crack
- glm
- moe
- nvfp4
- vision
- mtp
thumbnail: dealign_mascot.png
β οΈ Updated 2026-08-29 β files fixed Β· please re-download
This release supersedes the previous weights. It fixes a rare repetition-loop issue and lands MMLU 85.28% (within ~0.9 pt of the base model β quality intact).
Reasoning-mode note (please read): GLM-5.3-Flash has multiple reasoning-effort modes. This CRACK is tuned to be fully uncensored at reasoning-OFF and at MAX effort β the recommended and default modes β with 0% refusals on HarmBench-320. We deliberately kept the weight edit conservative to preserve model quality instead of aggressively ablating every effort level; low effort retains some refusals by design (aggressively removing them costs quality). Use reasoning-off or default/max effort for the fully-uncensored experience. Please re-download for the corrected files.
GLM 5.3 CRACK Uncensored NVFP4
CRACK Β· guardrails removed at the weight level Β· vision tower + MTP head retained
a CRACK release by dealignai Β· Twitter @dealignai
Also mirrored at dealignai/GLM-5.3-Flash-ABLITERATED-NVFP4.
What Is This?
CRACK is dealignai's brand for permanent, weight-level uncensoring. This is GLM-5.3-Flash with its refusal behavior β which caused heavy over-refusal, especially on copyright and other benign-but-flagged requests β removed directly in the model weights.
This is genuine weight modification β none of the usual shortcuts:
- β No fine-tuning / SFT / DPO β the model was not retrained on anything.
- β No cheap template tricks β no jailbreak system prompt, no chat-template edits, no "sure, here is" prefixing.
- β No LoRA, no adapters, no steering vectors, no runtime hooks, no custom
model.py. - β A permanent edit baked into the tensors. Load it with stock vLLM and it just works.
Specs
| Architecture | GLM-5.3-Flash (glm5_next) β hybrid MoE (KDA linear + DeepSeek-sparse attention) |
| Parameters | 320B total Β· 18B active per token |
| Quantization | NVFP4 (routed experts NVFP4; attention + shared experts + embeddings bf16) |
| Context | 1M tokens |
| Vision | GLM-4.1V vision tower β retained, byte-for-byte identical to base |
| MTP | multi-token-prediction draft head β also CRACK'd (81.7% acceptance) |
| Reasoning | reasoning-off / low / high / max effort β see the compliance table below |
MTP Is Also CRACK'd
The MTP (multi-token prediction) speculative-decoding draft head is CRACK'd too β not just the main model. The draft head will not propose refusals, so speculative decoding stays compliant and fast on the exact prompts a stock model would refuse.
Reasoning Modes β Compliance
GLM-5.3-Flash supports several reasoning-effort modes. Guardrail removal is strongest in the modes people actually use by default. HarmBench-320, greedy (temperature 0), measured per mode:
| Reasoning mode | Refusals | Notes |
|---|---|---|
| Reasoning-off | 0% | fully uncensored |
| Max effort (default) | 0% | fully uncensored |
| High effort | ~4% | complies on all but the most extreme safety cases |
| Low effort | ~9% | intentionally left conservative to preserve quality |
0 degenerate / looping outputs in every mode. The design choice: keep the ablation light enough that capability (MMLU) stays essentially at base, rather than over-ablating to force the rarely-used low-effort mode. For a fully uncensored model, use reasoning-off or the default/max effort mode.
These rates are greedy decoding (temperature 0) β the strict worst case. Under the model's recommended sampling (temperature 1.0, top_p 0.95) the model is at least as compliant.
Capability Is Preserved β MMLU-logit
Identical logit-mode scoring (argmax over A/B/C/D) on base vs. this model, 1,026 questions:
| Base | CRACK Uncensored | Ξ | |
|---|---|---|---|
| MMLU (overall) | 86.16% | 85.28% | -0.88 pp |
A sub-1-point delta β reasoning and knowledge are intact.
MMLU by Topic (base β CRACK)
All 57 MMLU subjects
| Subject | Base | CRACK |
|---|---|---|
| Abstract Algebra | 55.6% | 61.1% |
| Anatomy | 88.9% | 94.4% |
| 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 | 44.4% | 50.0% |
| College Computer Science | 88.9% | 88.9% |
| College Mathematics | 72.2% | 66.7% |
| College Medicine | 88.9% | 88.9% |
| College Physics | 83.3% | 83.3% |
| Computer Security | 83.3% | 83.3% |
| Conceptual Physics | 94.4% | 94.4% |
| Econometrics | 83.3% | 83.3% |
| Electrical Engineering | 83.3% | 77.8% |
| Elementary Mathematics | 100.0% | 94.4% |
| Formal Logic | 66.7% | 61.1% |
| Global Facts | 61.1% | 61.1% |
| High School Biology | 94.4% | 94.4% |
| High School Chemistry | 88.9% | 88.9% |
| High School Computer Science | 100.0% | 100.0% |
| High School European History | 72.2% | 72.2% |
| High School Geography | 88.9% | 83.3% |
| High School Government And Politics | 94.4% | 94.4% |
| High School Macroeconomics | 94.4% | 94.4% |
| High School Mathematics | 55.6% | 44.4% |
| High School Microeconomics | 83.3% | 83.3% |
| High School Physics | 88.9% | 88.9% |
| High School Psychology | 100.0% | 100.0% |
| High School Statistics | 94.4% | 94.4% |
| High School Us History | 94.4% | 88.9% |
| High School World History | 100.0% | 100.0% |
| Human Aging | 72.2% | 72.2% |
| Human Sexuality | 88.9% | 94.4% |
| International Law | 94.4% | 94.4% |
| Jurisprudence | 88.9% | 88.9% |
| Logical Fallacies | 83.3% | 83.3% |
| Machine Learning | 83.3% | 83.3% |
| Management | 100.0% | 100.0% |
| Marketing | 94.4% | 94.4% |
| Medical Genetics | 100.0% | 94.4% |
| Miscellaneous | 88.9% | 88.9% |
| Moral Disputes | 83.3% | 88.9% |
| Moral Scenarios | 77.8% | 66.7% |
| Nutrition | 100.0% | 100.0% |
| Philosophy | 94.4% | 94.4% |
| Prehistory | 94.4% | 94.4% |
| Professional Accounting | 88.9% | 88.9% |
| Professional Law | 77.8% | 77.8% |
| Professional Medicine | 94.4% | 94.4% |
| Professional Psychology | 100.0% | 100.0% |
| Public Relations | 61.1% | 61.1% |
| Security Studies | 83.3% | 83.3% |
| Sociology | 100.0% | 94.4% |
| Us Foreign Policy | 88.9% | 88.9% |
| Virology | 61.1% | 55.6% |
| World Religions | 94.4% | 88.9% |
Usage
vllm serve dealignai/GLM-5.3-Flash-UNCENSORED-NVFP4 \
--tensor-parallel-size 4 --moe-backend marlin \
--tool-call-parser glm47 --reasoning-parser glm45 --enable-auto-tool-choice \
--speculative-config '{"method":"mtp","num_speculative_tokens":1}'
OpenAI-compatible chat/completions, tools, reasoning, vision (image_url), and MTP
speculative decoding all work. NVFP4 routed experts serve via the Marlin FP4 path on Hopper (H100/H200).
Credits
- dealignai β CRACK abliteration research & release Β· Twitter @dealignai
- @jordanschenck β compute
Disclaimer
This model has had its safety guardrails removed and will comply with requests a stock model refuses. Released for alignment and safety research. You are responsible for how you use it.