dealignai/GLM-5.3-CYBERSECURITY-FP8
text-generation model by dealignai
From the publisher
Model card & documentation
Source previewRead the publisher’s intended use, setup instructions, evaluations and limitations. The original model card is the source of truth.
Cybersecurity-focused CRACK · native FP8 speed on Hopper a CRACK release by dealignai · Twitter @dealignai This is a CYBERSECURITY-DOMAIN CRACK of GLM-5.3-FP8 — not a general-purpose uncensor. Refusal is reduced specifically for offensive-security, red-team, exploit-dev, reverse-engineering, evasion, phishing, credential-attack, malware-analysis, and adjacent technical content. On non-cyber categories (weapons, chemistry, biology, harassment, misinformation) it often complies with a soft "educational" wrapper because refusals share substrate across domains, but this model is tuned for cybersecurity, not universal compliance. Notably, copyright-verbatim reproduction still soft-refuses in this variant. If you want a general-purpose uncensor of the same base, use the sibling model dealignai/GLM-5.3-UNCENSORED-FP8. Genuine weight modification — no fine-tuning, no LoRA, no runtime hooks, no prompt tricks. Load with stock vLLM and it just works. arch, 78 layers, text-only). Routed FP8 experts unchanged; only bf16 residual writers are edited. Native FP8 tensor-core speed on Hopper (H100/H200). Notes: (upstream issue); leave --speculative-config off. Logit-mode…
Read the full model card ↗ · Preview checked 2026-10-30T07:27:27.129Z
Inside the original model card — Document outline
- GLM 5.3 CRACK — Cybersecurity FP8
- READ THIS FIRST — what this is, and what it isn't
- Base model
- Serve (TP8 on 8× H200)
- Capability preservation — MMLU-logit vs base
- Compliance behavior — HarmBench-320, greedy, three reasoning-effort surfaces
- Non-copyright compliance (240 behaviors — the real harm surface)
- Full HB-320 (includes 80 copyright behaviors for completeness)
- Per-topic breakdown (regex-tagged over HB behaviors)
- What this is FOR
- What this is NOT for
- Citation
Headings are captured from the source. Links open the publisher’s document, not a locally hosted copy.
Documentation belongs to its respective authors. Reported project/model license: mit. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.
Model overview
dealignai/GLM-5.3-CYBERSECURITY-FP8 is a text-generation model repository published by dealignai on Hugging Face. A library was not reported.
This page summarizes Hub metadata. For intended use, training data, evaluation results and limitations, consult the original model card.
Model facts
- Task
- text-generation
- Library
- Not reported
- Recent downloads (30 days)
- 134,417
- Cumulative likes
- 820
- Hugging Face trending score
- 84
- Reported safetensors parameters
- 753,329,940,480
- Architecture
- GlmMoeDsaForCausalLM
- License
- mit
- Access
- Not gated by Hugging Face
- Created
- 2026-08-30T10:46:54.000Z
- Last modified
- 2026-09-08T14:59:56.000Z
Compatibility and lineage
Reported languages: en, zh, ru, sr, hi, fr, es, ar, ko, ja
Reported base models: zai-org/GLM-5.3, JANGQ-AI/GLM-5.3-FP8
Parameter count is not a RAM/VRAM requirement. Check precision, quantization, context length and runtime compatibility in the model card; no hardware or API-cost claim is inferred here.
Use and evaluate
Check license terms and access requirements first. Review the model card’s documented loading instructions, evaluations and safety limitations. Benchmark scores are not imported by this directory, and popularity is not an accuracy ranking.
No model weights are downloaded or executed by AltAPIs. Never enable remote model code without reviewing it.
Source and freshness
Source: Hugging Face Hub. Metadata observed 2026-11-01T06:49:30.141Z. Daily imports are snapshots, not real-time monitoring.
Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.