Instructions to use burtenshaw/Qwen3-30B-A3B-python-coder with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use burtenshaw/Qwen3-30B-A3B-python-coder with Transformers:

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

pipe = pipeline("text-generation", model="burtenshaw/Qwen3-30B-A3B-python-coder")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("burtenshaw/Qwen3-30B-A3B-python-coder")
model = AutoModelForCausalLM.from_pretrained("burtenshaw/Qwen3-30B-A3B-python-coder")
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 burtenshaw/Qwen3-30B-A3B-python-coder with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "burtenshaw/Qwen3-30B-A3B-python-coder"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "burtenshaw/Qwen3-30B-A3B-python-coder",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/burtenshaw/Qwen3-30B-A3B-python-coder

SGLang

How to use burtenshaw/Qwen3-30B-A3B-python-coder 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 "burtenshaw/Qwen3-30B-A3B-python-coder" \
    --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": "burtenshaw/Qwen3-30B-A3B-python-coder",
		"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 "burtenshaw/Qwen3-30B-A3B-python-coder" \
        --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": "burtenshaw/Qwen3-30B-A3B-python-coder",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use burtenshaw/Qwen3-30B-A3B-python-coder with Docker Model Runner:
```
docker model run hf.co/burtenshaw/Qwen3-30B-A3B-python-coder
```

burtenshaw HF Staff commited on May 2, 2025

Commit

81919af

verified ·

1 Parent(s): 02a8cd1

Upload train.py with huggingface_hub

Browse files

Files changed (1) hide show

train.py +1 -10

train.py CHANGED Viewed

@@ -11,7 +11,6 @@
 #     "trl",
 #     "peft",
 #     "wandb",
-#     "bitsandbytes",
 #     "torchvision",
 #     "torchaudio",
 # ]
@@ -22,7 +21,7 @@
 import torch
 from datasets import load_dataset
-from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
 from trl import SFTConfig, SFTTrainer, setup_chat_format
 from peft import LoraConfig
@@ -70,20 +69,12 @@ learning_rate = 2e-4
 """## Load model and tokenizer"""
-# specify how to quantize the model
-# quantization_config = BitsAndBytesConfig(
-#         load_in_4bit=True,
-#         bnb_4bit_quant_type="nf4",
-#         bnb_4bit_use_double_quant=True,
-# )
 # Load model
 model = AutoModelForCausalLM.from_pretrained(
     model_name,
     torch_dtype=torch.bfloat16,
     use_cache=False,  # Disable KV cache during training
     device_map="auto",
-    # quantization_config=quantization_config
 )
 # Load tokenizer

 #     "trl",
 #     "peft",
 #     "wandb",
 #     "torchvision",
 #     "torchaudio",
 # ]
 import torch
 from datasets import load_dataset
+from transformers import AutoModelForCausalLM, AutoTokenizer
 from trl import SFTConfig, SFTTrainer, setup_chat_format
 from peft import LoraConfig
 """## Load model and tokenizer"""
 # Load model
 model = AutoModelForCausalLM.from_pretrained(
     model_name,
     torch_dtype=torch.bfloat16,
     use_cache=False,  # Disable KV cache during training
     device_map="auto",
 )
 # Load tokenizer