Cyber-Tiel-Coder-35B-A3B
35BOtheruncensored
from huihui-ai/Huihui-Ornith-1.5-35B-A3B-abliterated
↓ 1M♥ 3433 repos
Gemory-26B-A4B
Gemory-26B-A4B is a refusal-ablated variant of UltimateIntent/Gemory-26B-A4B-GGUF, published on Hugging Face by UltimateIntent. It is 26B total parameters (mixture-of-experts), 48GB VRAM class class and apache-2.0 licence. It is distributed across 4 repositories in GGUF format, totalling 21.9K downloads.
Specification
- Model name
- Gemory-26B-A4B
- Source repository
- UltimateIntent/Gemory-26B-A4B-GGUF
- Base model
- UltimateIntent/Gemory-26B-A4B-GGUF
- Publisher
- UltimateIntent
- Parameters
- 26B (mixture-of-experts)
- VRAM required
- ~16.6 GB at Q4_K_M, 4k context (estimated) — runs on 24 GB cards
- 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
- 4 Sept 2026
- Last updated
- 14 Sept 2026
Ablation technique
How Gemory-26B-A4B 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
“A creative-writing and roleplay operative built to enter the scene, read the room, make the move, and keep going. Gemory carries 26B total parameters with roughly 4B active per token—a little femme fatale with very little interest in wasting motion.”
Running it
Commands are templates — confirm the exact repository and quant file before use.
llama.cpp / GGUF
hf download UltimateIntent/Gemory-26B-A4B-GGUF --local-dir ./gemory-26b-a4b
llama-cli -m ./gemory-26b-a4b/<file>.gguf -p "..."Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer
repo = "UltimateIntent/Gemory-26B-A4B-GGUF"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
repo, torch_dtype="auto", device_map="auto"
)vLLM
vllm serve UltimateIntent/Gemory-26B-A4B-GGUF --trust-remote-codeDownloads and variants (3)
Every published repository of Gemory-26B-A4B, including quantised re-releases by other authors.
| Repository | Format | Downloads |
|---|---|---|
| UltimateIntent/Gemory-26B-A4B-GGUFsource | GGUF | 17.1K |
| Abiray/Gemory-26B-A4B-GGUF | GGUF | 4.7K |
| DigitalBonsai/Gemory-26B-A4B-GGUF | GGUF | 75 |
Security-team assessment
Derived from this model’s metadata and the publisher’s claims — not from benchmarks run by this site.
- Gemory-26B-A4B 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 UltimateIntent/Gemory-26B-A4B-GGUF, not added by ablation — at 26B 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.
- Declared licence is apache-2.0, but the terms that bind you are UltimateIntent/Gemory-26B-A4B-GGUF’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 Gemory-26B-A4B is collected from the public Hugging Face API and the publisher’s model card. This site does not host weights, is not affiliated with UltimateIntent, and does not independently verify publisher claims. See responsible use.