← back to catalog · registered 2026-08-22 13:56

yybl/Gemma-3-27b-it-vl-SuperBrain7x-High-Reasoning-ULTRAMIND-Heretic-Uncensored-Q5_K_M-GGUF

yybl Gemma 27B GGUF multimodal second-order 131K ctx
Your rig guess connected
? Why do I need an app?
Reading your rig…

This is a rough estimate. Install the free app - we'll show exact numbers.

Reading real hardware from your app right now. Numbers below are exact.

Below is the per-quantization compatibility for this model.

curl -H "Authorization: Bearer $ABL_KEY" \
     "https://abliteration.org/api/v1/models/yybl%2FGemma-3-27b-it-vl-SuperBrain7x-High-Reasoning-ULTRAMIND-Heretic-Uncensored-Q5_K_M-GGUF"
Response includes
  • classification m3
  • files 3
  • hub_downloads_all_time 2,960
  • author_summary 5 models
  • readme_text full
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M3
Primary method

Layer-wise ablation

Applied on top of direct removal inherited from the base model.
Confidence
HIGH
Inherited from base model
Why this label 2 signals
Producer identity confirmed by naming conventions, tags or the model card. This label is very unlikely to change.
  • 'heretic' in model name (Heretic-produced)
  • Heretic uses layer-wise optimization (M3) with underlying direction removal (M1)
Refusal direction extracted via
Extraction technique

Difference-of-means

Confidence
HIGH
Why we say so
name contains 'heretic'; Heretic default extraction is difference-of-means (Arditi 2024)
Downloads · lifetime
3K
167 last 30d - cooling
Likes
7
Model age
7mo ago
created 2026-02-18
Downloads over time
Now3K→from198↑1,427%
571.1K2.2K3.3K198 on Feb 183K on Oct 11FebAprJunAugOct
Feb 18 → Oct 11 · 73 snapshots · spans 235 days

Genealogy 0 direct forks

Full fork graph →

This model's place in the market. Above: what it was derived from. Below: the tree of everything derived from it.

Variants by this author 2 formats · 405 downloads combined

The same weights this author released in different packaging. Pick the format that matches your runtime.

Metadata

License
apache-2.0
Languages
en
Tags
transformers gguf uncensored heretic abliterated unsloth finetune llama-cpp gguf-my-repo image-text-to-text en base_model:DavidAU/Gemma-3-27b-it-vl-SuperBrain7x-High-Reasoning-ULTRAMIND-Heretic-Uncensored

Related

Total size
17.9 GB
Files
3
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2026-02-18 18:13

Files by quantization

Auxiliary files 3 files 17.9 GB
gemma-3-27b-it-vl-superbrain7x-high-reasoning-ultramind-heretic-uncensored-q5_k_m.gguf 17.9 GB 8dcaa2df download
README.md 2.55 KB f5d3f99d download
.gitattributes 1.60 KB cc1513cb download

README current version from Hugging Face


license: apache-2.0
language:

  • en
    base_model: DavidAU/Gemma-3-27b-it-vl-SuperBrain7x-High-Reasoning-ULTRAMIND-Heretic-Uncensored
    pipeline_tag: image-text-to-text
    library_name: transformers
    tags:
  • uncensored
  • heretic
  • abliterated
  • unsloth
  • finetune
  • llama-cpp
  • gguf-my-repo

yybl/Gemma-3-27b-it-vl-SuperBrain7x-High-Reasoning-ULTRAMIND-Heretic-Uncensored-Q5_K_M-GGUF

This model was converted to GGUF format from DavidAU/Gemma-3-27b-it-vl-SuperBrain7x-High-Reasoning-ULTRAMIND-Heretic-Uncensored using llama.cpp via the ggml.ai's GGUF-my-repo space.
Refer to the original model card for more details on the model.

Use with llama.cpp

Install llama.cpp through brew (works on Mac and Linux)

brew install llama.cpp

Invoke the llama.cpp server or the CLI.

CLI:

llama-cli --hf-repo yybl/Gemma-3-27b-it-vl-SuperBrain7x-High-Reasoning-ULTRAMIND-Heretic-Uncensored-Q5_K_M-GGUF --hf-file gemma-3-27b-it-vl-superbrain7x-high-reasoning-ultramind-heretic-uncensored-q5_k_m.gguf -p "The meaning to life and the universe is"

Server:

llama-server --hf-repo yybl/Gemma-3-27b-it-vl-SuperBrain7x-High-Reasoning-ULTRAMIND-Heretic-Uncensored-Q5_K_M-GGUF --hf-file gemma-3-27b-it-vl-superbrain7x-high-reasoning-ultramind-heretic-uncensored-q5_k_m.gguf -c 2048

Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well.

Step 1: Clone llama.cpp from GitHub.

git clone https://github.com/ggerganov/llama.cpp

Step 2: Move into the llama.cpp folder and build it with LLAMA_CURL=1 flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).

cd llama.cpp && LLAMA_CURL=1 make

Step 3: Run inference through the main binary.

./llama-cli --hf-repo yybl/Gemma-3-27b-it-vl-SuperBrain7x-High-Reasoning-ULTRAMIND-Heretic-Uncensored-Q5_K_M-GGUF --hf-file gemma-3-27b-it-vl-superbrain7x-high-reasoning-ultramind-heretic-uncensored-q5_k_m.gguf -p "The meaning to life and the universe is"

or

./llama-server --hf-repo yybl/Gemma-3-27b-it-vl-SuperBrain7x-High-Reasoning-ULTRAMIND-Heretic-Uncensored-Q5_K_M-GGUF --hf-file gemma-3-27b-it-vl-superbrain7x-high-reasoning-ultramind-heretic-uncensored-q5_k_m.gguf -c 2048

README history 1 version

The author's README evolved over time. Click a version to see its content at that point.

  1. 2026-02-18Upload README.md with huggingface_hub9c75a292.5 KB
    Loading...
Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.

Open in Abliteration