Instructions to use maldv/Qwentile2.5-32B-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use maldv/Qwentile2.5-32B-Instruct with Transformers:

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

pipe = pipeline("text-generation", model="maldv/Qwentile2.5-32B-Instruct")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("maldv/Qwentile2.5-32B-Instruct")
model = AutoModelForCausalLM.from_pretrained("maldv/Qwentile2.5-32B-Instruct")
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

vLLM

How to use maldv/Qwentile2.5-32B-Instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "maldv/Qwentile2.5-32B-Instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "maldv/Qwentile2.5-32B-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/maldv/Qwentile2.5-32B-Instruct

SGLang

How to use maldv/Qwentile2.5-32B-Instruct 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 "maldv/Qwentile2.5-32B-Instruct" \
    --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": "maldv/Qwentile2.5-32B-Instruct",
		"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 "maldv/Qwentile2.5-32B-Instruct" \
        --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": "maldv/Qwentile2.5-32B-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use maldv/Qwentile2.5-32B-Instruct with Docker Model Runner:
```
docker model run hf.co/maldv/Qwentile2.5-32B-Instruct
```

More QwQ?

by mclassHF2023 - opened Dec 28, 2024

Discussion

mclassHF2023

Dec 28, 2024

I would be interested in merging in a bit more of the QwQ-preview and -abliterated models in the mix. It seems that for some logic puzzles, it doesn't perform nearly as good as more "QwQ-heavy" models.

Now, that might not have been the goal, and that's perfectly fine, I was just curious! :)

maldv

Owner Dec 29, 2024

Do you have an example of the problem you are trying to solve and how you are instructing it's solution?

maldv

Owner Jan 3, 2025

I refolded QwQ in again, with some other high quality models. Check out https://huggingface.co/maldv/Lytta2.5-32B-Instruct and let me know if it has better reasoning.

mclassHF2023

Jan 4, 2025

Hi, thanks for trying this out! Like you mention in its description, it is quite unhinged! :D
I will try it out some more, but I think it's more of a creative writing model and attempts of making it into a more "reasoning model" are probably wasted on it. But nevertheless it's an interesting result I think, I haven't seen any other model quite like it...

mclassHF2023

Jan 4, 2025

•

edited Jan 4, 2025

oh, and btw: one of the puzzles that I came up that even the best models like Gemini struggle with is this innocent looking question:

If you have one bucket that holds two gallons and another bucket that holds five gallons, how do you fill one of the buckets with exactly 4 gallons?

maldv

Owner Jan 6, 2025

I'm actually quite disappointed in it. I don't think I'm going to be able to improve on Qwentile without doing some sort of preference optimization. Thanks for trying it though.

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