Instructions to use Mushari440/Qwen3-8B-SFT-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Mushari440/Qwen3-8B-SFT-v2 with Transformers:

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

pipe = pipeline("text-generation", model="Mushari440/Qwen3-8B-SFT-v2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Mushari440/Qwen3-8B-SFT-v2")
model = AutoModelForCausalLM.from_pretrained("Mushari440/Qwen3-8B-SFT-v2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.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(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Mushari440/Qwen3-8B-SFT-v2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Mushari440/Qwen3-8B-SFT-v2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Mushari440/Qwen3-8B-SFT-v2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Mushari440/Qwen3-8B-SFT-v2

SGLang

How to use Mushari440/Qwen3-8B-SFT-v2 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 "Mushari440/Qwen3-8B-SFT-v2" \
    --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": "Mushari440/Qwen3-8B-SFT-v2",
		"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 "Mushari440/Qwen3-8B-SFT-v2" \
        --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": "Mushari440/Qwen3-8B-SFT-v2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Mushari440/Qwen3-8B-SFT-v2 with Docker Model Runner:
```
docker model run hf.co/Mushari440/Qwen3-8B-SFT-v2
```

Qwen3-8B-SFT

Model Details

Developed by: Mushari Alothman
Model type: Causal Language Model
Language(s): Arabic, English
License: Apache 2.0
Finetuned from: Qwen3-8B-Base

This is a supervised fine-tuned (SFT) Qwen3-8B model optimized for accurate, clean supervision across Arabic and English tasks.

Intended Uses

Direct Use

Arabic & English MCQ answering
Context-based QA / RAG
General instruction following

Out-of-Scope Use

Safety-critical or real-time decision making
Generating factual guarantees without verification

Training Summary

Training type: Supervised Fine-Tuning (SFT)
Precision: bf16 mixed precision
Data: Curated Arabic & English datasets including:
- MCQ
- QA / RAG / context understanding
- General instruction data

How to Use

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Mushari440/Qwen3-8B-SFT-v2")
model = AutoModelForCausalLM.from_pretrained("Mushari440/Qwen3-8B-SFT-v2")

inputs = tokenizer("سؤال: ما عاصمة السعودية؟", return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=50)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

Downloads last month: 550

Safetensors

Model size

8B params

Tensor type

BF16