Instructions to use 17Lab/qwen3b-lora-sft-s13 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use 17Lab/qwen3b-lora-sft-s13 with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2.5-3B")
model = PeftModel.from_pretrained(base_model, "17Lab/qwen3b-lora-sft-s13")

Transformers

How to use 17Lab/qwen3b-lora-sft-s13 with Transformers:

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

pipe = pipeline("text-generation", model="17Lab/qwen3b-lora-sft-s13")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("17Lab/qwen3b-lora-sft-s13", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use 17Lab/qwen3b-lora-sft-s13 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "17Lab/qwen3b-lora-sft-s13"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "17Lab/qwen3b-lora-sft-s13",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/17Lab/qwen3b-lora-sft-s13

SGLang

How to use 17Lab/qwen3b-lora-sft-s13 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 "17Lab/qwen3b-lora-sft-s13" \
    --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": "17Lab/qwen3b-lora-sft-s13",
		"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 "17Lab/qwen3b-lora-sft-s13" \
        --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": "17Lab/qwen3b-lora-sft-s13",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use 17Lab/qwen3b-lora-sft-s13 with Docker Model Runner:
```
docker model run hf.co/17Lab/qwen3b-lora-sft-s13
```

qwen3b-lora-sft-s13 / trainer_log.jsonl

Yiqi-Liu

Upload folder using huggingface_hub

fd099bb verified about 1 month ago

Raw

History Blame Contribute Delete

5.18 kB

	{"current_steps": 5, "total_steps": 114, "loss": 3.2117641448974608, "lr": 9.999999999999999e-05, "epoch": 0.13157894736842105, "percentage": 4.39, "elapsed_time": "0:00:04", "remaining_time": "0:01:27"}
	{"current_steps": 10, "total_steps": 114, "loss": 1.3626741409301757, "lr": 0.0001497146023568809, "epoch": 0.2631578947368421, "percentage": 8.77, "elapsed_time": "0:00:06", "remaining_time": "0:01:12"}
	{"current_steps": 15, "total_steps": 114, "loss": 0.37534074783325194, "lr": 0.00014797836529348678, "epoch": 0.39473684210526316, "percentage": 13.16, "elapsed_time": "0:00:09", "remaining_time": "0:01:05"}
	{"current_steps": 20, "total_steps": 114, "loss": 0.25732452869415284, "lr": 0.00014470106431701166, "epoch": 0.5263157894736842, "percentage": 17.54, "elapsed_time": "0:00:12", "remaining_time": "0:01:00"}
	{"current_steps": 25, "total_steps": 114, "loss": 0.22821950912475586, "lr": 0.0001399519052838329, "epoch": 0.6578947368421053, "percentage": 21.93, "elapsed_time": "0:00:15", "remaining_time": "0:00:56"}
	{"current_steps": 30, "total_steps": 114, "loss": 0.21986465454101561, "lr": 0.00013383117486896816, "epoch": 0.7894736842105263, "percentage": 26.32, "elapsed_time": "0:00:18", "remaining_time": "0:00:52"}
	{"current_steps": 35, "total_steps": 114, "loss": 0.19870233535766602, "lr": 0.000126468122840155, "epoch": 0.9210526315789473, "percentage": 30.7, "elapsed_time": "0:00:21", "remaining_time": "0:00:48"}
	{"current_steps": 40, "total_steps": 114, "loss": 0.18972501754760743, "lr": 0.00011801823272632844, "epoch": 1.0526315789473684, "percentage": 35.09, "elapsed_time": "0:00:24", "remaining_time": "0:00:45"}
	{"current_steps": 45, "total_steps": 114, "loss": 0.18694405555725097, "lr": 0.00010865993851503467, "epoch": 1.1842105263157894, "percentage": 39.47, "elapsed_time": "0:00:27", "remaining_time": "0:00:42"}
	{"current_steps": 50, "total_steps": 114, "loss": 0.18443360328674316, "lr": 9.859085671137102e-05, "epoch": 1.3157894736842106, "percentage": 43.86, "elapsed_time": "0:00:30", "remaining_time": "0:00:39"}
	{"current_steps": 50, "total_steps": 114, "eval_loss": 0.18096967041492462, "epoch": 1.3157894736842106, "percentage": 43.86, "elapsed_time": "0:00:31", "remaining_time": "0:00:39"}
	{"current_steps": 55, "total_steps": 114, "loss": 0.17949447631835938, "lr": 8.802361332501978e-05, "epoch": 1.4473684210526316, "percentage": 48.25, "elapsed_time": "0:00:39", "remaining_time": "0:00:42"}
	{"current_steps": 60, "total_steps": 114, "loss": 0.17545132637023925, "lr": 7.718135390573335e-05, "epoch": 1.5789473684210527, "percentage": 52.63, "elapsed_time": "0:00:42", "remaining_time": "0:00:38"}
	{"current_steps": 65, "total_steps": 114, "loss": 0.16337993144989013, "lr": 6.629303144060771e-05, "epoch": 1.7105263157894737, "percentage": 57.02, "elapsed_time": "0:00:45", "remaining_time": "0:00:34"}
	{"current_steps": 70, "total_steps": 114, "loss": 0.15300323963165283, "lr": 5.558857161731093e-05, "epoch": 1.8421052631578947, "percentage": 61.4, "elapsed_time": "0:00:48", "remaining_time": "0:00:30"}
	{"current_steps": 75, "total_steps": 114, "loss": 0.1532054901123047, "lr": 4.529401754706323e-05, "epoch": 1.973684210526316, "percentage": 65.79, "elapsed_time": "0:00:51", "remaining_time": "0:00:26"}
	{"current_steps": 80, "total_steps": 114, "loss": 0.1503889560699463, "lr": 3.562675647490799e-05, "epoch": 2.1052631578947367, "percentage": 70.18, "elapsed_time": "0:00:54", "remaining_time": "0:00:23"}
	{"current_steps": 85, "total_steps": 114, "loss": 0.15022125244140624, "lr": 2.6790929273509545e-05, "epoch": 2.236842105263158, "percentage": 74.56, "elapsed_time": "0:00:57", "remaining_time": "0:00:19"}
	{"current_steps": 90, "total_steps": 114, "loss": 0.15019209384918214, "lr": 1.897311965692531e-05, "epoch": 2.3684210526315788, "percentage": 78.95, "elapsed_time": "0:01:00", "remaining_time": "0:00:16"}
	{"current_steps": 95, "total_steps": 114, "loss": 0.15014156103134155, "lr": 1.2338414144029779e-05, "epoch": 2.5, "percentage": 83.33, "elapsed_time": "0:01:03", "remaining_time": "0:00:12"}
	{"current_steps": 100, "total_steps": 114, "loss": 0.15012743473052978, "lr": 7.026915972251254e-06, "epoch": 2.6315789473684212, "percentage": 87.72, "elapsed_time": "0:01:06", "remaining_time": "0:00:09"}
	{"current_steps": 100, "total_steps": 114, "eval_loss": 0.15011924505233765, "epoch": 2.6315789473684212, "percentage": 87.72, "elapsed_time": "0:01:06", "remaining_time": "0:00:09"}
	{"current_steps": 105, "total_steps": 114, "loss": 0.15010836124420165, "lr": 3.150786576338332e-06, "epoch": 2.763157894736842, "percentage": 92.11, "elapsed_time": "0:01:15", "remaining_time": "0:00:06"}
	{"current_steps": 110, "total_steps": 114, "loss": 0.1501091480255127, "lr": 7.918771064291645e-07, "epoch": 2.8947368421052633, "percentage": 96.49, "elapsed_time": "0:01:18", "remaining_time": "0:00:02"}
	{"current_steps": 114, "total_steps": 114, "eval_loss": 0.15010888874530792, "epoch": 3.0, "percentage": 100.0, "elapsed_time": "0:01:20", "remaining_time": "0:00:00"}
	{"current_steps": 114, "total_steps": 114, "epoch": 3.0, "percentage": 100.0, "elapsed_time": "0:01:25", "remaining_time": "0:00:00"}