Not-For-All-Audiences

nsfw

Instructions to use Undi95/PsyMedRP-v1-20B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Undi95/PsyMedRP-v1-20B with Transformers:

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

pipe = pipeline("text-generation", model="Undi95/PsyMedRP-v1-20B")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Undi95/PsyMedRP-v1-20B")
model = AutoModelForCausalLM.from_pretrained("Undi95/PsyMedRP-v1-20B")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Undi95/PsyMedRP-v1-20B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Undi95/PsyMedRP-v1-20B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Undi95/PsyMedRP-v1-20B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Undi95/PsyMedRP-v1-20B

SGLang

How to use Undi95/PsyMedRP-v1-20B 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 "Undi95/PsyMedRP-v1-20B" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Undi95/PsyMedRP-v1-20B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "Undi95/PsyMedRP-v1-20B" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Undi95/PsyMedRP-v1-20B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Undi95/PsyMedRP-v1-20B with Docker Model Runner:
```
docker model run hf.co/Undi95/PsyMedRP-v1-20B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

PsyMedRP-v1-13B-p1:
[jondurbin/airoboros-l2-13b-3.0](0.85) x [ehartford/Samantha-1.11-13b](0.15)

PsyMedRP-v1-13B-p2:
[Xwin-LM/Xwin-LM-13B-V0.1](0.85) x [chaoyi-wu/MedLLaMA_13B](0.15)

PsyMedRP-v1-20B-p1:
[PsyMedRP-v1-13B-p1](0.90) x [migtissera/Synthia-13B-v1.2](0.10)

PsyMedRP-v1-20B-p2:
[PsyMedRP-v1-13B-p2](0.90) x [migtissera/Synthia-13B-v1.2](0.10)

PsyMedRP-v1-20B-p3:
[Huginn merge with Gryphe gradient to PsyMedRP-v1-20B-p1]

PsyMedRP-v1-20B-p4:
[Huginn merge with Gryphe gradient to PsyMedRP-v1-20B-p2]

PsyMedRP-v1-20B-p5:
Apply Undi95/LimaRP-v3-120-Days at 0.3 weight to PsyMedRP-v1-20B-p3

PsyMedRP-v1-20B-p6:
Apply Undi95/LimaRP-v3-120-Days at 0.3 weight to PsyMedRP-v1-20B-p4

PsyMedRP-v1-20B:
layer_slices:
  - model: PsyMedRP-v1-20B-p5
    start: 0
    end: 16
  - model: PsyMedRP-v1-20B-p6
    start: 8
    end: 20
  - model: PsyMedRP-v1-20B-p5
    start: 17
    end: 32
  - model: PsyMedRP-v1-20B-p6
    start: 21
    end: 40

In testing.

If you want to support me, you can here.