Instructions to use thillaic/CBT-Copilot with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use thillaic/CBT-Copilot with Transformers:

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

pipe = pipeline("text-generation", model="thillaic/CBT-Copilot")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("thillaic/CBT-Copilot")
model = AutoModelForCausalLM.from_pretrained("thillaic/CBT-Copilot", device_map="auto")
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]:]))

PEFT
How to use thillaic/CBT-Copilot with PEFT:
```
Task type is invalid.
```
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use thillaic/CBT-Copilot with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/thillaic/CBT-Copilot

SGLang

How to use thillaic/CBT-Copilot 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 "thillaic/CBT-Copilot" \
    --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": "thillaic/CBT-Copilot",
		"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 "thillaic/CBT-Copilot" \
        --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": "thillaic/CBT-Copilot",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use thillaic/CBT-Copilot with Docker Model Runner:
```
docker model run hf.co/thillaic/CBT-Copilot
```

🧠 CBT-Copilot

CBT-Copilot is a fine-tuned version of meta-llama/Llama-3.2-3B-Instruct, designed to simulate compassionate and supportive dialogues in the style of Cognitive Behavioral Therapy (CBT).

Fine-tuned using LoRA on the Lumiiree/therapod-dpo dataset and merged into a standalone model, it supports deployment through transformers, vLLM, and other inference frameworks.

🚀 How to Use (with vLLM)

Serve this model using vLLM:

pip install vllm[serve]

python3 -m vllm.entrypoints.openai.api_server --model thillaic/CBT-Copilot

Then query it via the OpenAI-compatible API:

import openai

openai.api_key = "EMPTY"
openai.api_base = "http://localhost:8000/v1"

response = openai.ChatCompletion.create(
    model="CBT-Copilot",
    messages=[
        {"role": "system", "content": "You are a compassionate CBT therapist."},
        {"role": "user", "content": "I've been feeling really anxious lately. What can I do?"}
    ]
)

print(response["choices"][0]["message"]["content"])

🧠 Intended Use

This model is intended for:

Mental health chatbot research
Journaling and self-reflection tools
Prototyping conversational CBT agents

⚠️ Disclaimer: This model is not a replacement for licensed mental health professionals. It should only be used for educational, research, or prototyping purposes.

📜 License

Licensed under the MIT License.

🙏 Acknowledgements

Based on Meta’s LLaMA 3.2B Instruct model
Trained on Lumiiree/therapod-dpo
Fine-tuning performed with Hugging Face transformers, PEFT, and LoRA

🛠️ Model developed by Thillai Chithambaram

Downloads last month: 10

Safetensors

Model size

3B params

Tensor type

F32

Model tree for thillaic/CBT-Copilot

Base model

meta-llama/Llama-3.2-3B-Instruct

Adapter

(808)

this model

Adapters

2 models

thillaic
/

CBT-Copilot