Instructions to use Pinkstack/Parm-2-CoT-14B-16k-o1-QwQ with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Pinkstack/Parm-2-CoT-14B-16k-o1-QwQ with Transformers:

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

pipe = pipeline("text-generation", model="Pinkstack/Parm-2-CoT-14B-16k-o1-QwQ")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("Pinkstack/Parm-2-CoT-14B-16k-o1-QwQ")
model = AutoModelForMultimodalLM.from_pretrained("Pinkstack/Parm-2-CoT-14B-16k-o1-QwQ")
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 Settings

vLLM

How to use Pinkstack/Parm-2-CoT-14B-16k-o1-QwQ with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Pinkstack/Parm-2-CoT-14B-16k-o1-QwQ"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Pinkstack/Parm-2-CoT-14B-16k-o1-QwQ",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Pinkstack/Parm-2-CoT-14B-16k-o1-QwQ

SGLang

How to use Pinkstack/Parm-2-CoT-14B-16k-o1-QwQ 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 "Pinkstack/Parm-2-CoT-14B-16k-o1-QwQ" \
    --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": "Pinkstack/Parm-2-CoT-14B-16k-o1-QwQ",
		"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 "Pinkstack/Parm-2-CoT-14B-16k-o1-QwQ" \
        --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": "Pinkstack/Parm-2-CoT-14B-16k-o1-QwQ",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Pinkstack/Parm-2-CoT-14B-16k-o1-QwQ with Docker Model Runner:
```
docker model run hf.co/Pinkstack/Parm-2-CoT-14B-16k-o1-QwQ
```

Parm-2-CoT-14B-16k-o1-QwQ

File size: 1,015 Bytes

81554fc
25ddc0a
5e0d928
81554fc
 
25ddc0a
5e0d928
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
81554fc
 
d9fe9bd
 
5e0d928
f4ee896
25ddc0a
736b696
25ddc0a
b8979ff
f4ee896
81554fc
 
 
25ddc0a
f4ee896
81554fc
25ddc0a

---
license: mit
license_link: https://huggingface.co/microsoft/phi-4/resolve/main/LICENSE
language:
- en
pipeline_tag: text-generation
tags:
- phi
- nlp
- math
- code
- chat
- conversational
- phi3
inference:
  parameters:
    temperature: 0
widget:
- messages:
  - role: user
    content: How many R's in strawberry? Think step by step. 
library_name: transformers
---

gguf/final version: https://huggingface.co/Pinkstack/PARM-V2-phi-4-16k-CoT-o1-gguf

[Phi-4 Technical Report](https://arxiv.org/pdf/2412.08905)
Phi-4 that has been tuned to be more advanced at reasoning. Parm magic 😉

Unlike other Parm models we had to optimize our fine tuning process to ensure accuracy while still being able to release this model. **Training loss: 0.443800**

NOTE: more information soon, gguf

# Uploaded  model

- **Developed by:** Pinkstack
- **License:** MIT
- **Finetuned from model :** microsoft/phi-4

This phi-4 model was trained with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.