Instructions to use ameer4wisam/gemma-iraqi-10k-merged with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ameer4wisam/gemma-iraqi-10k-merged with Transformers:

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

pipe = pipeline("image-text-to-text", model="ameer4wisam/gemma-iraqi-10k-merged")
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, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("ameer4wisam/gemma-iraqi-10k-merged")
model = AutoModelForMultimodalLM.from_pretrained("ameer4wisam/gemma-iraqi-10k-merged", device_map="auto")
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 Settings

vLLM

How to use ameer4wisam/gemma-iraqi-10k-merged with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ameer4wisam/gemma-iraqi-10k-merged"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ameer4wisam/gemma-iraqi-10k-merged",
		"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/ameer4wisam/gemma-iraqi-10k-merged

SGLang

How to use ameer4wisam/gemma-iraqi-10k-merged 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 "ameer4wisam/gemma-iraqi-10k-merged" \
    --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": "ameer4wisam/gemma-iraqi-10k-merged",
		"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 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 "ameer4wisam/gemma-iraqi-10k-merged" \
        --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": "ameer4wisam/gemma-iraqi-10k-merged",
		"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"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use ameer4wisam/gemma-iraqi-10k-merged with Docker Model Runner:
```
docker model run hf.co/ameer4wisam/gemma-iraqi-10k-merged
```

Gemma 4 12B — اللهجة العراقية (مدموج)

نموذج كامل مدموج: محوّل LoRA مدرَّب على اللهجة العراقية مدموج داخل google/gemma-4-12B-it. ما يحتاج تحميل محوّل منفصل ولا مكتبة peft.

الدمج انسوى بـAutoModelForImageTextToText مو AutoModelForCausalLM، حتى تنحفظ أبراج الرؤية والصوت — الدمج بالكلاس النصي يفقدها.


النموذج الأساسي	`google/gemma-4-12B-it`
المحوّل المصدر	`ameer4wisam/gemma-iraqi-10k`
بيانات التدريب	14,000 مثال عراقي + ~15% تعليمات عامة (Aya)
الوسائط	نص + صورة (محفوظة بعد الدمج)

الاستخدام

import torch
from transformers import AutoProcessor, AutoModelForImageTextToText

model_id = "ameer4wisam/gemma-iraqi-10k-merged"
model = AutoModelForImageTextToText.from_pretrained(
    model_id, dtype=torch.bfloat16, device_map="auto"
)
processor = AutoProcessor.from_pretrained(model_id)

msgs = [
    {"role": "system", "content": "أنت مساعد عراقي تحچي باللهجة العراقية العامية."},
    {"role": "user", "content": "شلونك؟ شنو أشهر أكلة عراقية؟"},
]
inputs = processor.apply_chat_template(
    msgs, add_generation_prompt=True, tokenize=True,
    return_dict=True, return_tensors="pt"
).to(model.device)

out = model.generate(**inputs, max_new_tokens=64, do_sample=False)
print(processor.decode(out[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True))

سلوك الوكيل

بيانات التدريب فيها 690 مثالاً تعلّم سلوك مساعد مبيعات: الإحالة (اتأكدلك وأرد عليك) بدل اختراع الضمان والأسعار، رفض البراند غير المتوفر باسمه، تأجيل الحساب، علامة [ORDER_READY] بعد التأكيد الصريح حصراً، واستدعاء الأدوات بصيغة [TOOL_CALL].

هذي السلوكيات تعتمد على رسالة نظام تحدد الكتالوج والقواعد.

الحدود

اللهجة العراقية مو موحّدة؛ الداتا تميل للهجة الوسط والجنوب.
الأسعار والمنتجات بأمثلة التدريب توضيحية، مو كتالوج حقيقي.
ما انقاس على معيار عام منشور؛ التقييم داخلي بدفتر التدريب.

المصدر: https://github.com/ameer20042005/iraqi_words_finetuning

Downloads last month: 55

Safetensors

Model size

12B params

Tensor type

BF16

Model tree for ameer4wisam/gemma-iraqi-10k-merged

Base model

google/gemma-4-12B

Finetuned

google/gemma-4-12B-it

Finetuned

(130)

this model