rbrisita/gemma-4-26b-a4b-it-mm

rbrisita/ gemma-4-26b-a4b-it-mm

55 Downloads Updated 2 weeks ago

Gemma 4 26B A4B IT MM QAT Unsloth Dynamic 2.0 GGUF MoE

vision tools thinking

ollama run rbrisita/gemma-4-26b-a4b-it-mm:qat-ud-q4_k_xl

curl http://localhost:11434/api/chat \
  -d '{
    "model": "rbrisita/gemma-4-26b-a4b-it-mm:qat-ud-q4_k_xl",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='rbrisita/gemma-4-26b-a4b-it-mm:qat-ud-q4_k_xl',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'rbrisita/gemma-4-26b-a4b-it-mm:qat-ud-q4_k_xl',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Applications

Claude Code ollama launch claude --model rbrisita/gemma-4-26b-a4b-it-mm:qat-ud-q4_k_xl

OpenCode ollama launch opencode --model rbrisita/gemma-4-26b-a4b-it-mm:qat-ud-q4_k_xl

Hermes Agent ollama launch hermes --model rbrisita/gemma-4-26b-a4b-it-mm:qat-ud-q4_k_xl

OpenClaw ollama launch openclaw --model rbrisita/gemma-4-26b-a4b-it-mm:qat-ud-q4_k_xl

Models

View all →

Name

1 model

Size / Usage

Context

Input

gemma-4-26b-a4b-it-mm:qat-ud-q4_k_xl

16GB · 256K context window · Text, Image · 2 weeks ago

gemma-4-26b-a4b-it-mm:qat-ud-q4_k_xl

16GB

256K

Text, Image

Readme

Instruction-Tuned Multi-Modal Mixture-of-Experts with 4B Active Parameters. Quantization-Aware Training (QAT) reduces memory requirements while preserving model quality. This model contains a draft model that can be turned off by setting draft_num_predict=0. It also contains a multimodal projector (mmproj) with F16 weights.

Modelfile parameters taken from:

Unsloth Gemma 4

Supporting Details:

Tested on:

 OS: Ubuntu 26.04 resolute
 Kernel: x86_64 Linux 7.0.0-30-generic
 Shell: bash 5.3.9
 Disk: 1TB
 CPU: 12th Gen Intel Core i7-12700H @ 20x 4.6GHz
 RAM: 64GB 
 GPU: NVIDIA GeForce RTX 3050 Laptop GPU
 GPU RAM: 4GB