Gemma 4 Uncensored 26B (IQ4_XS) - Optimized for 16GB VRAM

tools thinking

ollama run VladimirGav/gemma4-26b-16GB-VRAM-Uncensored

curl http://localhost:11434/api/chat \
  -d '{
    "model": "VladimirGav/gemma4-26b-16GB-VRAM-Uncensored",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='VladimirGav/gemma4-26b-16GB-VRAM-Uncensored',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'VladimirGav/gemma4-26b-16GB-VRAM-Uncensored',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Applications

Claude Code ollama launch claude --model VladimirGav/gemma4-26b-16GB-VRAM-Uncensored

OpenClaw ollama launch openclaw --model VladimirGav/gemma4-26b-16GB-VRAM-Uncensored

Hermes Agent ollama launch hermes --model VladimirGav/gemma4-26b-16GB-VRAM-Uncensored

Codex ollama launch codex --model VladimirGav/gemma4-26b-16GB-VRAM-Uncensored

OpenCode ollama launch opencode --model VladimirGav/gemma4-26b-16GB-VRAM-Uncensored

Models

View all →

Name

1 model

Size / Usage

Context

Input

gemma4-26b-16GB-VRAM-Uncensored:latest

14GB · 256K context window · Text · 2 weeks ago

gemma4-26b-16GB-VRAM-Uncensored:latest

14GB

256K

Text

Readme

Gemma 4 26B Uncensored (IQ4_XS) - Optimized for 16GB VRAM Uncensored

This is a highly optimized, Uncensored version of Google Gemma 4 26B, specifically tailored to run on GPUs with 16GB of VRAM. It uses advanced IQ4_XS (Importance Matrix) quantization to maintain high intelligence and creativity while fitting into a limited memory footprint.

🚀 Key Features

Uncensored: No pre-installed filters, moralizing, or refusal to answer. Perfect for roleplay, unrestricted creative writing, and deep research.
VRAM Efficient: Occupies ~15GB, leaving about 1GB for context (KV Cache) on a 16GB card.

💻 Target Hardware

Perfectly fits: * NVIDIA RTX 5060 Ti (16GB) * NVIDIA RTX 4070 Ti Super (16GB) * NVIDIA RTX 3080 (16GB version) * NVIDIA RTX 4080 / 4080 Super * NVIDIA RTX ⁵⁰⁰⁰⁄₆₀₀₀ Ada / A4000 (16GB)

🛠 How to Use

Simply run the following command in your terminal:

ollama run VladimirGav/gemma4-26b-uncensored-16GB-VRAM