Instructions to use royallab/PsyOrca2-13b-DARE-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use royallab/PsyOrca2-13b-DARE-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="royallab/PsyOrca2-13b-DARE-GGUF",
	filename="PsyOrca-2-DARE-13b-gguf.q4_K_S.gguf",
)

output = llm(
	"Once upon a time,",
	max_tokens=512,
	echo=True
)
print(output)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use royallab/PsyOrca2-13b-DARE-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf royallab/PsyOrca2-13b-DARE-GGUF:Q4_K_S
# Run inference directly in the terminal:
llama cli -hf royallab/PsyOrca2-13b-DARE-GGUF:Q4_K_S

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf royallab/PsyOrca2-13b-DARE-GGUF:Q4_K_S
# Run inference directly in the terminal:
llama cli -hf royallab/PsyOrca2-13b-DARE-GGUF:Q4_K_S

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf royallab/PsyOrca2-13b-DARE-GGUF:Q4_K_S
# Run inference directly in the terminal:
./llama-cli -hf royallab/PsyOrca2-13b-DARE-GGUF:Q4_K_S

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf royallab/PsyOrca2-13b-DARE-GGUF:Q4_K_S
# Run inference directly in the terminal:
./build/bin/llama-cli -hf royallab/PsyOrca2-13b-DARE-GGUF:Q4_K_S

Use Docker

docker model run hf.co/royallab/PsyOrca2-13b-DARE-GGUF:Q4_K_S

LM Studio
Jan
Ollama
How to use royallab/PsyOrca2-13b-DARE-GGUF with Ollama:
```
ollama run hf.co/royallab/PsyOrca2-13b-DARE-GGUF:Q4_K_S
```

Unsloth Studio

How to use royallab/PsyOrca2-13b-DARE-GGUF with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for royallab/PsyOrca2-13b-DARE-GGUF to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for royallab/PsyOrca2-13b-DARE-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for royallab/PsyOrca2-13b-DARE-GGUF to start chatting

Atomic Chat new
Docker Model Runner
How to use royallab/PsyOrca2-13b-DARE-GGUF with Docker Model Runner:
```
docker model run hf.co/royallab/PsyOrca2-13b-DARE-GGUF:Q4_K_S
```

Lemonade

How to use royallab/PsyOrca2-13b-DARE-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull royallab/PsyOrca2-13b-DARE-GGUF:Q4_K_S

Run and chat with the model

lemonade run user.PsyOrca2-13b-DARE-GGUF-Q4_K_S

List all available models

lemonade list

PsyOrca2-DARE-13b-GGUF

This is a public GGUF test of PsyOrca2-DARE-13b. If this model is favorable, the FP16 will be uploaded with more details.

The model details are below:

Model

This is a Llama 2-based model consisting of a merge between:

KoboldAI/PsyFighter-2-13b (FP16 not available to the public yet. However, the merge config is.)
microsoft/Orca-2-13b

The goal of this merge is to test out the DARE merge algorithm and see how it works with these two models.

Mergekit config (Inspired from Charles Goddard):

models:
  - model: KoboldAI/Psyfighter-2-13B
    parameters:
      weight: 1
      density: 1
  - model: microsoft/Orca-2-13b
    parameters:
      weight: 0.05
      density: 0.30
merge_method: dare_ties
base_model: meta-llama/Llama-2-13b-hf
parameters:
  int8_mask: true
dtype: bfloat16

Usage:

The intended prompt format is unknown, but it can be safely assumed that it is compatible with Alpaca and Orca ChatML formatting.

Bias, Risks, and Limitations

The model will show biases similar to those observed in niche roleplaying forums on the Internet, besides those exhibited by the base model. It is not intended for supplying factual information or advice in any form.

Training Details

This model is a merge. Please refer to the link repositories of the merged models for details.

Downloads last month: 7

GGUF

Model size

13B params

Architecture

llama

Hardware compatibility

4-bit

5-bit

6-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support