Instructions to use Psychotherapy-LLM/PsyCoPref-Llama3-8B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Psychotherapy-LLM/PsyCoPref-Llama3-8B with Transformers:

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

pipe = pipeline("text-generation", model="Psychotherapy-LLM/PsyCoPref-Llama3-8B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Psychotherapy-LLM/PsyCoPref-Llama3-8B")
model = AutoModelForCausalLM.from_pretrained("Psychotherapy-LLM/PsyCoPref-Llama3-8B")
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 Psychotherapy-LLM/PsyCoPref-Llama3-8B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Psychotherapy-LLM/PsyCoPref-Llama3-8B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Psychotherapy-LLM/PsyCoPref-Llama3-8B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Psychotherapy-LLM/PsyCoPref-Llama3-8B

SGLang

How to use Psychotherapy-LLM/PsyCoPref-Llama3-8B 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 "Psychotherapy-LLM/PsyCoPref-Llama3-8B" \
    --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": "Psychotherapy-LLM/PsyCoPref-Llama3-8B",
		"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 "Psychotherapy-LLM/PsyCoPref-Llama3-8B" \
        --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": "Psychotherapy-LLM/PsyCoPref-Llama3-8B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Psychotherapy-LLM/PsyCoPref-Llama3-8B with Docker Model Runner:
```
docker model run hf.co/Psychotherapy-LLM/PsyCoPref-Llama3-8B
```

Responses to questions not performing optimally

by jeffreylubowski - opened Mar 17, 2025

Discussion

jeffreylubowski

Mar 17, 2025

Dear all, has anyone else come across issues while using this model? My question is around getting answers or responses that come back with random thoughts, not even referring to the base LLM for a generalist response (which would be much better). Even when I used a QA set from the dataset this model was fine-tuned on, the response was nowhere near what the training reflected.

I would love to talk to someone who worked on the fine-tuning, maybe I'm doing something wrong.

Thank you

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