Instructions to use AscendKernelGen/KernelGen-LM-32B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use AscendKernelGen/KernelGen-LM-32B with Transformers:

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

pipe = pipeline("text-generation", model="AscendKernelGen/KernelGen-LM-32B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("AscendKernelGen/KernelGen-LM-32B")
model = AutoModelForCausalLM.from_pretrained("AscendKernelGen/KernelGen-LM-32B")
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 AscendKernelGen/KernelGen-LM-32B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "AscendKernelGen/KernelGen-LM-32B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "AscendKernelGen/KernelGen-LM-32B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/AscendKernelGen/KernelGen-LM-32B

SGLang

How to use AscendKernelGen/KernelGen-LM-32B 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 "AscendKernelGen/KernelGen-LM-32B" \
    --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": "AscendKernelGen/KernelGen-LM-32B",
		"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 "AscendKernelGen/KernelGen-LM-32B" \
        --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": "AscendKernelGen/KernelGen-LM-32B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use AscendKernelGen/KernelGen-LM-32B with Docker Model Runner:
```
docker model run hf.co/AscendKernelGen/KernelGen-LM-32B
```

AscendKernelGen commited on Mar 13

Commit

2a6428e

verified ·

1 Parent(s): 2bfaa94

Update README.md

Browse files

Files changed (1) hide show

README.md +0 -8

README.md CHANGED Viewed

@@ -25,11 +25,3 @@ Our framework, **AscendKernelGen (AKGen)**, bridges the gap between general-purp
 * **Hardware-Grounded Evaluation:** Validated using **NPUKernelBench**, a comprehensive benchmark that assesses compilation success, functional correctness, and performance (latency) on real Ascend hardware across varying complexity levels.
 * **Performance:** The model demonstrates siginificant improvement on complex Level-2 kernels compared to baselines, and effectively solving tasks where general-purpose models (like Qwen3, Llama3.1) fail completely.
-## Citation
-@article{cao2026ascendkernelgen,
-  title={AscendKernelGen: A Systematic Study of LLM-Based Kernel Generation for Neural Processing Units},
-  author={Xinzi Cao and Jianyang Zhai and Pengfei Li and Zhiheng Hu and Cen Yan and Bingxu Mu and Guanghuan Fang and Bin She and Jiayu Li and Yihan Su and Dongyang Tao and Xiansong Huang and Fan Xu and Feidiao Yang and Yao Lu and Chang-Dong Wang and Yutong Lu and Weicheng Xue and Bin Zhou and Yonghong Tian},
-  journal={arXiv preprint arXiv:2601.07160},
-  year={2026},
-  url={https://arxiv.org/abs/2601.07160}
-}


25	* Hardware-Grounded Evaluation: Validated using NPUKernelBench, a comprehensive benchmark that assesses compilation success, functional correctness, and performance (latency) on real Ascend hardware across varying complexity levels.
26	* Performance: The model demonstrates siginificant improvement on complex Level-2 kernels compared to baselines, and effectively solving tasks where general-purpose models (like Qwen3, Llama3.1) fail completely.
27