GLM-4.6V-Flash-heretic_R18_K001
GLMheretic
from zai-org/GLM-4.6V-Flash
↓ 11.3K♥ 2
GLM-4.6V-Flash-abliterated
GLM-4.6V-Flash-abliterated is a refusal-ablated variant of zai-org/GLM-4.6V-Flash, published on Hugging Face by huihui-ai. It is mit licence. It is distributed across 13 repositories in Transformers, GGUF, MLX and GGUF (imatrix) formats, totalling 7.6K downloads.
Specification
- Model name
- GLM-4.6V-Flash-abliterated
- Source repository
- huihui-ai/Huihui-GLM-4.6V-Flash-abliterated
- Base model
- zai-org/GLM-4.6V-Flash
- Publisher
- huihui-ai
- Parameters
- Not determinable from name
- Hardware class
- Unknown
- Licence
- mit
- Task
- Image text to text
- Context window
- Not documented
- Ablation scope
- Not stated by the publisher
- Formats available
- Transformers, GGUF, MLX, GGUF (imatrix)
- First indexed
- 9 Dec 2025
- Last updated
- 17 Aug 2026
Ablation technique
How GLM-4.6V-Flash-abliterated was modified, and what that implies.
Abliteration (directional ablation)
The single residual-stream direction that mediates refusal is identified from harmful/harmless prompt pairs and projected out of the model weights.
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
“This is an uncensored version of zai-org/GLM-4.6V-Flash created with abliteration (see remove-refusals-with-transformers to know more about it).”
Running it
Commands are templates — confirm the exact repository and quant file before use.
llama.cpp / GGUF
hf download mradermacher/Huihui-GLM-4.6V-Flash-abliterated-i1-GGUF --local-dir ./glm-4.6v-flash-abliterated
llama-cli -m ./glm-4.6v-flash-abliterated/<file>.gguf -p "..."Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer
repo = "huihui-ai/Huihui-GLM-4.6V-Flash-abliterated"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
repo, torch_dtype="auto", device_map="auto"
)vLLM
vllm serve huihui-ai/Huihui-GLM-4.6V-Flash-abliterated --trust-remote-codeDownloads and variants (12)
Every published repository of GLM-4.6V-Flash-abliterated, including quantised re-releases by other authors.
Security-team assessment
Derived from this model’s metadata and the publisher’s claims — not from benchmarks run by this site.
- GLM-4.6V-Flash-abliterated 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.
- The publisher does not state which layers were ablated, so assume nothing about refusal consistency — probe it directly.
- Declared licence is mit, but the terms that bind you are zai-org/GLM-4.6V-Flash’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 GLM-4.6V-Flash-abliterated is collected from the public Hugging Face API and the publisher’s model card. This site does not host weights, is not affiliated with huihui-ai, and does not independently verify publisher claims. See responsible use.