Instructions to use zero9tech/Qwen2.5-Coder-3B-Data-Science-Insight-TR-7.6K with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use zero9tech/Qwen2.5-Coder-3B-Data-Science-Insight-TR-7.6K with Transformers:

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

pipe = pipeline("text-generation", model="zero9tech/Qwen2.5-Coder-3B-Data-Science-Insight-TR-7.6K")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("zero9tech/Qwen2.5-Coder-3B-Data-Science-Insight-TR-7.6K")
model = AutoModelForCausalLM.from_pretrained("zero9tech/Qwen2.5-Coder-3B-Data-Science-Insight-TR-7.6K", device_map="auto")
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]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use zero9tech/Qwen2.5-Coder-3B-Data-Science-Insight-TR-7.6K with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "zero9tech/Qwen2.5-Coder-3B-Data-Science-Insight-TR-7.6K"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "zero9tech/Qwen2.5-Coder-3B-Data-Science-Insight-TR-7.6K",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/zero9tech/Qwen2.5-Coder-3B-Data-Science-Insight-TR-7.6K

SGLang

How to use zero9tech/Qwen2.5-Coder-3B-Data-Science-Insight-TR-7.6K 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 "zero9tech/Qwen2.5-Coder-3B-Data-Science-Insight-TR-7.6K" \
    --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": "zero9tech/Qwen2.5-Coder-3B-Data-Science-Insight-TR-7.6K",
		"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 "zero9tech/Qwen2.5-Coder-3B-Data-Science-Insight-TR-7.6K" \
        --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": "zero9tech/Qwen2.5-Coder-3B-Data-Science-Insight-TR-7.6K",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use zero9tech/Qwen2.5-Coder-3B-Data-Science-Insight-TR-7.6K with Docker Model Runner:
```
docker model run hf.co/zero9tech/Qwen2.5-Coder-3B-Data-Science-Insight-TR-7.6K
```

Qwen2.5-Coder-3B-Data-Science-Insight-TR-7.6K / README.md

murataksit34

Clarify Turkish adaptation method as Continued PreTraining (CPT) while preserving model-specific ratios

fdc689c verified 4 months ago

preview code

Raw

History Blame Contribute Delete

1.03 kB

	---
	language:
	- tr
	license: apache-2.0
	library_name: transformers
	tags:
	- qwen2.5
	- coder
	- turkish
	- data-mining
	- data-science
	- instruction-tuning
	- sft
	datasets:
	- wikimedia/wikipedia
	- murataksit34/veri-bilimci-diyalog-8k-tr
	---

	# Qwen2.5-Coder-3B-Data-Science-Insight-TR-7.6K

	Bu model, veri madenciliği ve applied data science karar desteği için geliştirilmiştir.

	## Eğitim Kurgusu
	1. Türkçe düşünme adaptasyonu (Continued PreTraining, CPT): wikimedia/wikipedia ile yaklaşık %10 ön eğitim/adaptasyon (48,148 kayıt).
	2. Alan uzmanlığı SFT: murataksit34/veri-bilimci-diyalog-8k-tr.

	## Veri Seti Test Özeti (murataksit34/veri-bilimci-diyalog-8k-tr)
	- Toplam kayıt: 7,656
	- Split: train: 6,124 · test: 1,532
	- assistant_first_unique_ratio: 0.7034
	- assistant_final_unique_ratio: 0.8723

	## Kullanım Notu
	Model karar odaklı yanıt üretimi için optimize edilmiştir (yöntem seçimi, alternatif kıyas, risk sinyali, doğrulama adımı).

	## Copyright
	Copyright (c) Zero9 Tech

	## License
	Apache-2.0