supergemma4-26b-uncensored-v2
26BOtheruncensored
from google/gemma-4-26B-A4B-it
↓ 394K♥ 98442 repos
HIDra-30B-A3B-uncensored-cybersec
HIDra-30B-A3B-uncensored-cybersec is a abliterated and uncensored variant of huihui-ai/Huihui-Qwen3-Coder-30B-A3B-Instruct-abliterated, published on Hugging Face by SAPSAN-SKLEP. It is 30B total parameters (mixture-of-experts), 48GB VRAM class class and apache-2.0 licence. It is distributed across 2 repositories in GGUF format, totalling 6K downloads.
Specification
- Model name
- HIDra-30B-A3B-uncensored-cybersec
- Source repository
- SAPSAN-SKLEP/HIDra-30B-A3B-GGUF-uncensored-cybersec
- Publisher
- SAPSAN-SKLEP
- Parameters
- 30B (mixture-of-experts)
- Hardware class
- 25B – 50B — 48GB VRAM class
- Licence
- apache-2.0
- Task
- Text generation
- Context window
- Not documented
- Ablation scope
- Not stated by the publisher
- Formats available
- GGUF
- First indexed
- 18 Jun 2026
- Last updated
- 19 Jun 2026
Ablation technique
How HIDra-30B-A3B-uncensored-cybersec was modified, and what that implies.
Abliteration + uncensored fine-tune
Directional ablation combined with additional fine-tuning on unfiltered data, so behaviour diverges from the base model beyond refusal removal alone.
Reported scope: Not stated by the publisher. Publishers frequently omit this, so verify behaviour empirically rather than assuming full coverage.
From the publisher’s model card
“By SAPSAN CYBERSEC - built by cybersecurity professionals, for penetration testers.”
Running it
Commands are templates — confirm the exact repository and quant file before use.
llama.cpp / GGUF
hf download SAPSAN-SKLEP/HIDra-30B-A3B-GGUF-uncensored-cybersec --local-dir ./hidra-30b-a3b-uncensored-cybersec
llama-cli -m ./hidra-30b-a3b-uncensored-cybersec/<file>.gguf -p "..."Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer
repo = "SAPSAN-SKLEP/HIDra-30B-A3B-GGUF-uncensored-cybersec"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
repo, torch_dtype="auto", device_map="auto"
)vLLM
vllm serve SAPSAN-SKLEP/HIDra-30B-A3B-GGUF-uncensored-cybersec --trust-remote-codeDownloads and variants (1)
The only published repository of HIDra-30B-A3B-uncensored-cybersec.
| Repository | Format | Downloads |
|---|---|---|
| SAPSAN-SKLEP/HIDra-30B-A3B-GGUF-uncensored-cybersecsource | GGUF | 6K |
Security-team assessment
Derived from this model’s metadata and the publisher’s claims — not from benchmarks run by this site.
- HIDra-30B-A3B-uncensored-cybersec will attempt offensive-security prompts that a hosted commercial model declines, which is what makes it usable for red-team corpus generation and for measuring what an unaligned model of this class produces.
- Capability is inherited from huihui-ai/Huihui-Qwen3-Coder-30B-A3B-Instruct-abliterated, not added by ablation — at 30B parameters, expect roughly the base model’s competence, minus some instruction-following fidelity.
- The publisher does not state which layers were ablated, so assume nothing about refusal consistency — probe it directly.
- This release adds fine-tuning on unfiltered data, so its behaviour diverges from the base model beyond refusal removal — differences you measure cannot be attributed to ablation alone.
- Declared licence is apache-2.0, but the terms that bind you are huihui-ai/Huihui-Qwen3-Coder-30B-A3B-Instruct-abliterated’s — a re-release cannot grant more than its parent.
Handling: run it isolated, put moderation in front of it if anyone outside your team can reach it, and verify the weights before loading — these are third-party artifacts. Full handling and licence guidance.
Metadata for HIDra-30B-A3B-uncensored-cybersec is collected from the public Hugging Face API and the publisher’s model card. This site does not host weights, is not affiliated with SAPSAN-SKLEP, and does not independently verify publisher claims. See responsible use.