Instructions to use esokullu/gemma4-turkish-26b-a4b-pruned with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use esokullu/gemma4-turkish-26b-a4b-pruned with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="esokullu/gemma4-turkish-26b-a4b-pruned")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForImageTextToText

processor = AutoProcessor.from_pretrained("esokullu/gemma4-turkish-26b-a4b-pruned")
model = AutoModelForImageTextToText.from_pretrained("esokullu/gemma4-turkish-26b-a4b-pruned")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use esokullu/gemma4-turkish-26b-a4b-pruned with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "esokullu/gemma4-turkish-26b-a4b-pruned"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "esokullu/gemma4-turkish-26b-a4b-pruned",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/esokullu/gemma4-turkish-26b-a4b-pruned

SGLang

How to use esokullu/gemma4-turkish-26b-a4b-pruned with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "esokullu/gemma4-turkish-26b-a4b-pruned" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "esokullu/gemma4-turkish-26b-a4b-pruned",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "esokullu/gemma4-turkish-26b-a4b-pruned" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "esokullu/gemma4-turkish-26b-a4b-pruned",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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

Gemma 4 26B-A4B Pruned + Türkçe Heal (bf16)

google/gemma-4-26B-A4B-it modelinin uzman-budanmış (expert-pruned) ve Türkçeye uyarlanmış bf16 versiyonu. Her katmandaki 128 uzmandan 101'i korundu (~%21 küçülme), ardından Türkçe + kod + matematik karışık talimat veriseti üzerinde LoRA ile iyileştirildi.

GGUF (llama.cpp / Ollama / LM Studio) versiyonu için: gemma4-turkish-26b-a4b-pruned-gguf

Donanım Gereksinimleri

bf16 inference: ~40 GB VRAM (A100, H100, 2× RTX 5090, vb.)
Tek tüketici kart için 4-bit quantized GGUF sürümünü öneriyoruz

Kullanım

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model_id = "esokullu/gemma4-tr-26b-a4b-pruned"
tok = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id, torch_dtype=torch.bfloat16, device_map="auto"
)

msgs = [
    {"role": "system", "content": "Sen Türkçe konuşan yardımsever bir asistansın."},
    {"role": "user",   "content": "Kuantum dolanıklığını basitçe açıklar mısın?"},
]
enc = tok.apply_chat_template(msgs, return_tensors="pt", add_generation_prompt=True,
                              return_dict=True).to(model.device)
out = model.generate(**enc, max_new_tokens=400, do_sample=True,
                     temperature=0.7, top_p=0.9, repetition_penalty=1.1)
print(tok.decode(out[0][enc["input_ids"].shape[-1]:], skip_special_tokens=True))

Eğitim Detayları

Aşama	Detay
Calibration	6000 örnek (%50 Türkçe + %25 kod + %25 matematik), router aktivasyonları loglandı
Prune	K=101 (128'den), routing weight × count metriğiyle, fused 3D expert + `router.per_expert_scale` aux sliced
Heal	LoRA r=32 α=64, 2 epoch, 25k örnek (%60 TR + %20 kod + %20 matematik), all-linear targets
Donanım	A100 80GB, ~2 saat

Sınırlamalar

Matematik akıl yürütme (çok adımlı problemler) zayıf
Türkçe-dışı dillerde yazılı talimatlara varsayılan olarak Türkçe cevap verir

Lisans

Gemma kullanım koşullarına tabidir. Google Gemma Terms of Use

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 mix.

For consumer GPUs use the GGUF release.

Training Summary

Stage	Detail
Calibration	6000 samples, 50% Turkish + 25% code + 25% math
Prune	K=101 per layer, routing-weight × count, fused 3D expert + per_expert_scale aux
Heal	LoRA r=32 α=64, 2 epochs, 25k samples (60% TR + 20% code + 20% math)
Hardware	A100 80GB, ~2h total

License

Subject to Gemma Terms of Use.

Downloads last month: 110

Safetensors

Model size

21B params

Tensor type

BF16

Model tree for esokullu/gemma4-turkish-26b-a4b-pruned

Base model

google/gemma-4-26B-A4B

Finetuned

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

Finetuned

(95)

this model