Instructions to use kyLELEng/retailops-instruct-qwen3.5-9b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use kyLELEng/retailops-instruct-qwen3.5-9b with Transformers:

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

pipe = pipeline("text-generation", model="kyLELEng/retailops-instruct-qwen3.5-9b")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("kyLELEng/retailops-instruct-qwen3.5-9b")
model = AutoModelForCausalLM.from_pretrained("kyLELEng/retailops-instruct-qwen3.5-9b")
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

vLLM

How to use kyLELEng/retailops-instruct-qwen3.5-9b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "kyLELEng/retailops-instruct-qwen3.5-9b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "kyLELEng/retailops-instruct-qwen3.5-9b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/kyLELEng/retailops-instruct-qwen3.5-9b

SGLang

How to use kyLELEng/retailops-instruct-qwen3.5-9b 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 "kyLELEng/retailops-instruct-qwen3.5-9b" \
    --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": "kyLELEng/retailops-instruct-qwen3.5-9b",
		"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 "kyLELEng/retailops-instruct-qwen3.5-9b" \
        --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": "kyLELEng/retailops-instruct-qwen3.5-9b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use kyLELEng/retailops-instruct-qwen3.5-9b with Docker Model Runner:
```
docker model run hf.co/kyLELEng/retailops-instruct-qwen3.5-9b
```

retailops-instruct-qwen3.5-9b / training_config.json

kyLELEng

Upload merged RetailOps-Instruct Qwen3.5-9B full model

a750b19 verified 24 days ago

raw

history blame contribute delete

1.08 kB

	{
	"base_model_id": "Qwen/Qwen3.5-9B",
	"dataset_repo_id": "kyLELEng/retailops-instruct-sft",
	"model_repo_id": "kyLELEng/retailops-instruct-qwen3.5-9b-lora",
	"output_dir": "/tmp/retailops-instruct-qwen35-9b-lora",
	"smoke_test": false,
	"force_rebuild_dataset": false,
	"dataset_only": false,
	"max_total_examples": 300000,
	"max_ec_examples": 70000,
	"max_shopify_examples": 40000,
	"max_review_examples": 170000,
	"max_c4_examples": 20000,
	"review_scan_per_category": 180000,
	"meta_scan_per_category": 120000,
	"max_steps": 600,
	"max_length": 2048,
	"per_device_train_batch_size": 1,
	"per_device_eval_batch_size": 1,
	"gradient_accumulation_steps": 2,
	"learning_rate": 0.0001,
	"warmup_ratio": 0.03,
	"weight_decay": 0.01,
	"logging_steps": 10,
	"eval_steps": 100,
	"save_steps": 300,
	"save_total_limit": 2,
	"dataset_num_proc": 8,
	"lora_r": 64,
	"lora_alpha": 128,
	"lora_dropout": 0.05,
	"seed": 42,
	"mixed_precision": "bf16",
	"attn_implementation": "sdpa",
	"packing": false,
	"eval_generation_samples": 8,
	"max_new_tokens": 800
	}