Instructions to use webbrain-one/gemma4-turkish-26b-a4b-pruned-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 webbrain-one/gemma4-turkish-26b-a4b-pruned-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 webbrain-one/gemma4-turkish-26b-a4b-pruned-gguf:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf webbrain-one/gemma4-turkish-26b-a4b-pruned-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 webbrain-one/gemma4-turkish-26b-a4b-pruned-gguf:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf webbrain-one/gemma4-turkish-26b-a4b-pruned-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 webbrain-one/gemma4-turkish-26b-a4b-pruned-gguf:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf webbrain-one/gemma4-turkish-26b-a4b-pruned-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 webbrain-one/gemma4-turkish-26b-a4b-pruned-gguf:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf webbrain-one/gemma4-turkish-26b-a4b-pruned-gguf:Q4_K_M

Use Docker

docker model run hf.co/webbrain-one/gemma4-turkish-26b-a4b-pruned-gguf:Q4_K_M

Llama new
LM Studio
Jan

vLLM

How to use webbrain-one/gemma4-turkish-26b-a4b-pruned-gguf with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "webbrain-one/gemma4-turkish-26b-a4b-pruned-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": "webbrain-one/gemma4-turkish-26b-a4b-pruned-gguf",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/webbrain-one/gemma4-turkish-26b-a4b-pruned-gguf:Q4_K_M

Ollama
How to use webbrain-one/gemma4-turkish-26b-a4b-pruned-gguf with Ollama:
```
ollama run hf.co/webbrain-one/gemma4-turkish-26b-a4b-pruned-gguf:Q4_K_M
```
Unsloth Desktop

How to use webbrain-one/gemma4-turkish-26b-a4b-pruned-gguf with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf webbrain-one/gemma4-turkish-26b-a4b-pruned-gguf:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "webbrain-one/gemma4-turkish-26b-a4b-pruned-gguf:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use webbrain-one/gemma4-turkish-26b-a4b-pruned-gguf with Docker Model Runner:
```
docker model run hf.co/webbrain-one/gemma4-turkish-26b-a4b-pruned-gguf:Q4_K_M
```

Lemonade

How to use webbrain-one/gemma4-turkish-26b-a4b-pruned-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull webbrain-one/gemma4-turkish-26b-a4b-pruned-gguf:Q4_K_M

Run and chat with the model

lemonade run user.gemma4-turkish-26b-a4b-pruned-gguf-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use webbrain-one/gemma4-turkish-26b-a4b-pruned-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 webbrain-one/gemma4-turkish-26b-a4b-pruned-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 webbrain-one/gemma4-turkish-26b-a4b-pruned-gguf:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use webbrain-one/gemma4-turkish-26b-a4b-pruned-gguf with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf webbrain-one/gemma4-turkish-26b-a4b-pruned-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 "webbrain-one/gemma4-turkish-26b-a4b-pruned-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"

Gemma 4 26B-A4B Pruned + Türkçe Heal — GGUF

google/gemma-4-26B-A4B-it modelinin uzman-budanmış (expert-pruned) ve Türkçeye uyarlanmış versiyonu. Her katmandaki 128 uzmandan 101'i korundu (~%21 küçülme), ardından Türkçe + kod + matematik karışık talimat veriseti üzerinde LoRA ile iyileştirildi.

llama.cpp / Ollama / LM Studio ile RTX 5060 (16 GB) veya benzeri tüketici GPU'larda kolayca çalışır.

Sürümler

Dosya	Boyut	Açıklama
`merged.Q4_K_M.gguf`	~11 GB	Önerilen — 24 GB+ VRAM'li kartlarda hızlı
`merged.Q5_K_M.gguf`	~13 GB	Daha yüksek kalite, biraz daha yavaş

Kullanım

llama.cpp

./llama-cli -m merged.Q4_K_M.gguf \
  -p "Türkiye'nin başkenti neresidir?" \
  -n 200 --temp 0.7 --repeat-penalty 1.1

Ollama

cat > Modelfile <<'MODELFILE'
FROM ./merged.Q4_K_M.gguf
TEMPLATE """{{ if .System }}<start_of_turn>system
{{ .System }}<end_of_turn>
{{ end }}{{ if .Prompt }}<start_of_turn>user
{{ .Prompt }}<end_of_turn>
<start_of_turn>model
{{ end }}{{ .Response }}<end_of_turn>
"""
PARAMETER stop "<end_of_turn>"
PARAMETER repeat_penalty 1.1
SYSTEM "Sen yardımsever, dürüst ve bilgili bir Türkçe konuşan asistansın. Kullanıcının sorusu hangi dilde olursa olsun her zaman Türkçe yanıt ver."
MODELFILE

ollama create gemma-tr -f Modelfile
ollama run gemma-tr

Eğitim Detayları

Aşama	Detay
Calibration	6000 örnek, %50 Türkçe + %25 kod + %25 matematik karışım
Prune	K=101 (128'den), routing weight × count metriğiyle skor
Heal	LoRA r=32 α=64, 2 epoch, 25k örnek (%60 TR + %20 kod + %20 matematik)
Donanım	A100 80GB, ~2 saat

Sınırlamalar

Matematik akıl yürütme (özellikle çok adımlı problemler) zayıf
4-bit quant nedeniyle bf16 versiyonuna kıyasla ~%1-2 perplexity artışı
Türkçe-dışı dillerde yazılı talimatlara hâlâ Türkçe cevap verir (system prompt ile değiştirilebilir)

Gemma 4 26B-A4B Pruned + Turkish Heal — GGUF (English)

Expert-pruned variant of google/gemma-4-26B-A4B-it: 128 → 101 experts per layer (~21% smaller), then LoRA-healed on a Turkish + code + math instruction mix. Runs comfortably on RTX 5090 / 24 GB+ consumer GPUs via llama.cpp / Ollama / LM Studio.

Variants

File	Size	Notes
`merged.Q4_K_M.gguf`	~11 GB	Recommended for 24 GB+ VRAM
`merged.Q5_K_M.gguf`	~13 GB	Higher quality, still fits

Training summary

Stage	Detail
Calibration	6000 samples, 50% Turkish + 25% code + 25% math
Prune	K=101 per layer, ranked by routing weight × count
Heal	LoRA r=32 α=64, 2 epochs, 25k samples (60% TR + 20% code + 20% math)
Hardware	A100 80GB, ~2 hours

Limitations

Multi-step math reasoning is weak
~1-2% perplexity increase vs bf16 due to 4-bit quantization
Defaults to Turkish output; override via system prompt for other languages

Downloads last month: 245

GGUF

Model size

20B params

Architecture

gemma4

Hardware compatibility

2-bit

3-bit

4-bit

5-bit

16-bit

Model tree for webbrain-one/gemma4-turkish-26b-a4b-pruned-gguf

Base model

google/gemma-4-26B-A4B

Finetuned

google/gemma-4-26B-A4B-it

Quantized

(414)

this model

Collection including webbrain-one/gemma4-turkish-26b-a4b-pruned-gguf

Turkish Language Models & Datasets

Collection

3 items • Updated about 7 hours ago