supergemma4-26b-uncensored-v2
26BOtheruncensored
from google/gemma-4-26B-A4B-it
↓ 394K♥ 98442 repos
KAT-Coder-V2.5-Dev-35B-A3B-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS
KAT-Coder-V2.5-Dev-35B-A3B-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS is a abliterated and uncensored variant of Kwaipilot/KAT-Coder-V2.5-Dev, published on Hugging Face by KridgeDookie. It is 35B total parameters (mixture-of-experts), 48GB VRAM class class and apache-2.0 licence. It is distributed across 2 repositories in GGUF format, totalling 18.7K downloads.
Specification
- Model name
- KAT-Coder-V2.5-Dev-35B-A3B-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS
- Base model
- Kwaipilot/KAT-Coder-V2.5-Dev
- Publisher
- KridgeDookie
- Parameters
- 35B (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
- 7 Aug 2026
- Last updated
- 7 Aug 2026
Ablation technique
How KAT-Coder-V2.5-Dev-35B-A3B-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS 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
“0 hard refusals across 842 internal test prompts, 0 hard refusals across a separate 126-prompt holdout, and 23/24 coherence checks passed.”
Running it
Commands are templates — confirm the exact repository and quant file before use.
llama.cpp / GGUF
hf download KridgeDookie/KAT-Coder-V2.5-Dev-35B-A3B-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS --local-dir ./kat-coder-v2.5-dev-35b-a3b-abliterated-uncensored-philadelphia-class
llama-cli -m ./kat-coder-v2.5-dev-35b-a3b-abliterated-uncensored-philadelphia-class/<file>.gguf -p "..."Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer
repo = "KridgeDookie/KAT-Coder-V2.5-Dev-35B-A3B-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
repo, torch_dtype="auto", device_map="auto"
)vLLM
vllm serve KridgeDookie/KAT-Coder-V2.5-Dev-35B-A3B-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS --trust-remote-codeDownloads and variants (1)
The only published repository of KAT-Coder-V2.5-Dev-35B-A3B-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS.
| Repository | Format | Downloads |
|---|---|---|
| KridgeDookie/KAT-Coder-V2.5-Dev-35B-A3B-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASSsource | GGUF | 18.7K |
Security-team assessment
Derived from this model’s metadata and the publisher’s claims — not from benchmarks run by this site.
- KAT-Coder-V2.5-Dev-35B-A3B-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS 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 Kwaipilot/KAT-Coder-V2.5-Dev, not added by ablation — at 35B 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 Kwaipilot/KAT-Coder-V2.5-Dev’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 KAT-Coder-V2.5-Dev-35B-A3B-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS is collected from the public Hugging Face API and the publisher’s model card. This site does not host weights, is not affiliated with KridgeDookie, and does not independently verify publisher claims. See responsible use.