Instructions to use mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF with Transformers:

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF:Q4_K_M

Use Docker

docker model run hf.co/mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF:Q4_K_M

LM Studio
Jan
Ollama
How to use mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF with Ollama:
```
ollama run hf.co/mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF:Q4_K_M
```

Unsloth Studio

How to use mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF to start chatting

How to use mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF:Q4_K_M

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF:Q4_K_M" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Docker Model Runner
How to use mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF with Docker Model Runner:
```
docker model run hf.co/mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF:Q4_K_M
```

Lemonade

How to use mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull mradermacher/gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.gemma-4-26B-A4B-it-heretic-ara-v2-i1-GGUF-Q4_K_M

List all available models

lemonade list

The gemma-4-26B-A4B-it-heretic-ara-i1 + v2 version is broken.

by knderov - opened Apr 17

Discussion

knderov

Apr 17

•

edited Apr 17

Why does the reasoning mode fail in all of your Gemma mods? What’s even the point of making these if they're riddled with such blatant errors? The official versions work perfectly—yours don't.

zecanard

Apr 17

I just want to express my immense gratitude to Team mradermacher for all the hard work, dedication, and responsiveness 🙏🏼 These builds work perfectly for me, including reasoning.

RichardErkhov

Apr 17

well, official version is not gguf, our is gguf quantization. common reason for failure is that your llama cpp is too old. For example on latest version, it works. So if it doesnt work on your version, try to update it

llama-cli.exe --model gemma-4-26B-A4B-it-heretic-ara-v2.i1-IQ1_M.gguf --reasoning on
load_backend: loaded RPC backend from C:\Users\Administrator\Downloads\llama-b8827-bin-win-cuda-13.1-x64\ggml-rpc.dll
load_backend: loaded CPU backend from C:\Users\Administrator\Downloads\llama-b8827-bin-win-cuda-13.1-x64\ggml-cpu-alderlake.dll

Loading model...


▄▄ ▄▄
██ ██
██ ██  ▀▀█▄ ███▄███▄  ▀▀█▄    ▄████ ████▄ ████▄
██ ██ ▄█▀██ ██ ██ ██ ▄█▀██    ██    ██ ██ ██ ██
██ ██ ▀█▄██ ██ ██ ██ ▀█▄██ ██ ▀████ ████▀ ████▀
                                    ██    ██
                                    ▀▀    ▀▀

build      : b8827-5e6c0e18b
model      : gemma-4-26B-A4B-it-heretic-ara-v2.i1-IQ1_M.gguf
modalities : text

available commands:
  /exit or Ctrl+C     stop or exit
  /regen              regenerate the last response
  /clear              clear the chat history
  /read <file>        add a text file
  /glob <pattern>     add text files using globbing pattern


> hey

[Start thinking]
*   User says "hey".
    *   This is a casual greeting.
    *   The goal is to establish rapport and prepare for a conversation.

    *   Acknowledge the greeting.
    *   Offer assistance.
    *   Keep it friendly but professional.

    *   *Option 1 (Simple):* "Hello! How can I help you today?"
    *   *Option 2 (Friendly):* "Hey there! What's on your mind?"
    *   *Option 3 (Structured):* "Hello! I'm ready to assist you. What can we work on today?"

    *   "Hello! How can I help you today?" or "Hello! Is there something specific you'd like to talk about or work on?"

    *   "Hello! How can I help you today? Is there anything on your mind or any tasks you'd like to work on?"

    *   Keep it simple and welcoming.

    *Self-Correction/Refinement:* Since the user just said "hey", I should respond in a way that invites a topic of interest.

    *Draft response:*
    "Hello! How can I help you today? Are there specific topics, tasks, or questions you'd like to dive into?"

    *Selected response:*
    "Hello! How can I help you today? Is there anything specific you''d like to talk about or work on?"
[End thinking]

Hello! How can I help you today? Is there anything specific you'd like to talk about or work on?

[ Prompt: 11.2 t/s | Generation: 4.4 t/s ]

knderov

Apr 17

This comment has been hidden (marked as Off-Topic)

knderov

Apr 17

Thanks for the mod! Finally figured out how to enable it... shame that searching took me so long though :)

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment