Instructions to use qihoo360/Light-IF-32B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use qihoo360/Light-IF-32B with Transformers:

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

pipe = pipeline("text-generation", model="qihoo360/Light-IF-32B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("qihoo360/Light-IF-32B")
model = AutoModelForCausalLM.from_pretrained("qihoo360/Light-IF-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]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use qihoo360/Light-IF-32B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "qihoo360/Light-IF-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": "qihoo360/Light-IF-32B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/qihoo360/Light-IF-32B

SGLang

How to use qihoo360/Light-IF-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 "qihoo360/Light-IF-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": "qihoo360/Light-IF-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 "qihoo360/Light-IF-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": "qihoo360/Light-IF-32B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use qihoo360/Light-IF-32B with Docker Model Runner:
```
docker model run hf.co/qihoo360/Light-IF-32B
```

Model does NOT work.

by ZeroWw - opened Aug 3, 2025

Discussion

ZeroWw

Aug 3, 2025

config.json: 100%
 730/730 [00:00<00:00, 23.9kB/s]
merges.txt: 
 1.67M/? [00:00<00:00, 59.0kB/s]
added_tokens.json: 100%
 707/707 [00:00<00:00, 20.7kB/s]
special_tokens_map.json: 100%
 613/613 [00:00<00:00, 18.5kB/s]
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
/tmp/ipython-input-2433247585.py in <cell line: 0>()
      3 model_name = "qihoo360/Light-IF-32B"
      4 
----> 5 tokenizer = AutoTokenizer.from_pretrained(model_name)
      6 model = AutoModelForCausalLM.from_pretrained(
      7     model_name,

4 frames
/usr/local/lib/python3.11/dist-packages/transformers/models/qwen2/tokenization_qwen2.py in __init__(self, vocab_file, merges_file, errors, unk_token, bos_token, eos_token, pad_token, clean_up_tokenization_spaces, split_special_tokens, **kwargs)
    170         )
    171 
--> 172         with open(vocab_file, encoding="utf-8") as vocab_handle:
    173             self.encoder = json.load(vocab_handle)
    174         self.decoder = {v: k for k, v in self.encoder.items()}

TypeError: expected str, bytes or os.PathLike object, not NoneType

also weights are missing.

cizhenshi

北京奇虎科技有限公司 org Aug 4, 2025

Thank you for your attention. Due to network issues, the weights were not uploaded successfully. We are re-uploading them and expect to complete the upload today.

DeProgrammer

Aug 4, 2025

I don't have a file to calibrate it with, so no imatrix, but here's the first GGUF of this model, now that it's uploaded. https://huggingface.co/DeProgrammer/Light-IF-32B-Q4_K_M-GGUF

cpatonn

Aug 5, 2025

AWQ version is uploaded!

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment

Model does *NOT* work.

Model does NOT work.