Instructions to use keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-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 keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-GGUF:F16
# Run inference directly in the terminal:
llama cli -hf keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-GGUF:F16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-GGUF:F16
# Run inference directly in the terminal:
llama cli -hf keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-GGUF:F16

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 keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-GGUF:F16
# Run inference directly in the terminal:
./llama-cli -hf keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-GGUF:F16

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 keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-GGUF:F16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-GGUF:F16

Use Docker

docker model run hf.co/keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-GGUF:F16

LM Studio
Jan

vLLM

How to use keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-GGUF",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-GGUF:F16

Ollama
How to use keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-GGUF with Ollama:
```
ollama run hf.co/keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-GGUF:F16
```

Unsloth Studio

How to use keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-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 keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-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 keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-GGUF to start chatting

How to use keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-GGUF:F16

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": "keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-GGUF:F16"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

OpenClaw new

How to use keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-GGUF:F16

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 "keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-GGUF:F16" \
  --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 keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-GGUF with Docker Model Runner:
```
docker model run hf.co/keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-GGUF:F16
```

Lemonade

How to use keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-GGUF:F16

Run and chat with the model

lemonade run user.Gemma-4-26B-A4B-it-MixQ-13G-GGUF-F16

List all available models

lemonade list

Hermes Agent

How to use keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-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 keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-GGUF:F16

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 keyuan01/Gemma-4-26B-A4B-it-MixQ-13G-GGUF:F16

Run Hermes

hermes

Atomic Chat

keyuan01 commited on Apr 12

Commit

8013662

verified ·

1 Parent(s): 801e6ae

Upload README.md with huggingface_hub

Browse files

Files changed (1) hide show

README.md +15 -13

README.md CHANGED Viewed

@@ -51,19 +51,21 @@ This is not a single uniform `Q4` or `Q3` file.
 It is a mixed-precision build where different tensor groups keep different quant types according to sensitivity and size budget.
 ## Mix Formula
-The published 13GB file is centered around this mix behavior:
-- main attention and core blocks are mostly kept in `Q6_K`
-- a large secondary block remains in `Q4_K`
-- selected tensors fall back to `Q5_0` and `Q4_0`
-- only a small compressed tail reaches `Q3_K`
-- one tensor remains in `Q5_K`
-Practical summary:
-- `Q6_K` is the main high-preservation part
-- `Q4_K` is the main compression body
-- `Q3_K` is only a small minority, not the center of the file
-So this release should not be described as a pure `Q3_K_M`, `IQ3_M`, or pure `Q4_K_M` build.
 ## Importance Matrix (`imatrix`)
 This release follows the same `imatrix`-guided quantization idea used in the 31B line.
@@ -137,4 +139,4 @@ License:
 Use this version if you want:
 - a published mixed 13GB GGUF for `Gemma 4 26B A4B it`
 - multimodal support preserved through the separate `mmproj`
-- a file whose actual mix is closer to `Q6_K + Q4_K` than to a pure `Q3` build

 It is a mixed-precision build where different tensor groups keep different quant types according to sensitivity and size budget.
 ## Mix Formula
+This published 13GB file follows a mixed recipe in this style:
+- `token_embd -> q5_k`
+- `output -> q5_k`
+- `router -> q8_0`
+- `attn_q -> q6_k`
+- `attn_k -> q6_k`
+- `attn_v -> q6_k`
+- `attn_output -> q6_k`
+- `ffn_gate_up_exps -> mixed q4_k / q3_k`
+- `ffn_down_exps -> q4_0`
+Notes:
+- this file is MoE, so expert tensors are not laid out like the dense 31B recipe
+- `ffn_gate_up_exps` is the main mixed expert block
+- the 13GB release is therefore closer to a `Q6_K + Q4_K/Q3_K` expert mix than to a `Q3-centered` dense recipe
 ## Importance Matrix (`imatrix`)
 This release follows the same `imatrix`-guided quantization idea used in the 31B line.
 Use this version if you want:
 - a published mixed 13GB GGUF for `Gemma 4 26B A4B it`
 - multimodal support preserved through the separate `mmproj`
+- a MoE mixed quant release documented in recipe style instead of a generic quant summary