Instructions to use TheBloke/openchat_3.5-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use TheBloke/openchat_3.5-GGUF with Transformers:

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("TheBloke/openchat_3.5-GGUF", dtype="auto")

llama-cpp-python

How to use TheBloke/openchat_3.5-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="TheBloke/openchat_3.5-GGUF",
	filename="openchat_3.5.Q2_K.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 TheBloke/openchat_3.5-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf TheBloke/openchat_3.5-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf TheBloke/openchat_3.5-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf TheBloke/openchat_3.5-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf TheBloke/openchat_3.5-GGUF:Q4_K_M

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 TheBloke/openchat_3.5-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf TheBloke/openchat_3.5-GGUF:Q4_K_M

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 TheBloke/openchat_3.5-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf TheBloke/openchat_3.5-GGUF:Q4_K_M

Use Docker

docker model run hf.co/TheBloke/openchat_3.5-GGUF:Q4_K_M

LM Studio
Jan
Ollama
How to use TheBloke/openchat_3.5-GGUF with Ollama:
```
ollama run hf.co/TheBloke/openchat_3.5-GGUF:Q4_K_M
```

Unsloth Studio

How to use TheBloke/openchat_3.5-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 TheBloke/openchat_3.5-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 TheBloke/openchat_3.5-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for TheBloke/openchat_3.5-GGUF to start chatting

Atomic Chat new
Docker Model Runner
How to use TheBloke/openchat_3.5-GGUF with Docker Model Runner:
```
docker model run hf.co/TheBloke/openchat_3.5-GGUF:Q4_K_M
```

Lemonade

How to use TheBloke/openchat_3.5-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull TheBloke/openchat_3.5-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.openchat_3.5-GGUF-Q4_K_M

List all available models

lemonade list

I really need help with Templates? How to interpret and use them?

by Goldenblood56 - opened Nov 4, 2023

Discussion

Goldenblood56

Nov 4, 2023

•

edited Nov 4, 2023

I use Oobabooga. In the readme it says...

prompt_template: |
GPT4 User: {prompt}<|end_of_turn|>GPT4 Assistant:

How do I use this in instruct mode in Oobabooga? Or do I have to use chat mode? I prefer having a long discussion with the AI not role play but asking questions and getting it to write stories based on what we discussed etc. Kind of like how one would use ChatGPT. Basically how can I add that template into Ooba? I've been to the Instruct template tab. I know how to change templates. But not how to properly format or add new ones etc.

There is an OpenChat template in Oobabooga

User string
GPT4 User:

Bot String
GPT4 Assistant:

Turn template
<|user|> <|user-message|><|end_of_turn|><|bot|> <|bot-message|><|end_of_turn|>

Command for chat-instrruct mode
Continue the chat dialogue below. Write a single reply for the character "<|character|>".

<|prompt|>

But this is not the correct one right? How do I use the one in the readme?

piotr-sikora

Nov 14, 2023

I have same question. How to use System in template?

@Goldenblood56 did you figure out?

Goldenblood56

Nov 15, 2023

No I did not.

mufeed

Nov 26, 2023

I have same question. How to use System in template?

@Goldenblood56 did you figure out?

Try:
Turn template

<|user|> <|user-message|><|end_of_turn|><|bot|> <|bot-message|>

User string

GPT4 User:

Bot string

GPT4 Assistant:

Goldenblood56

Dec 13, 2023

Thank you. I think I stopped using this model but the information provided may still help me. I might try this out. I notice for a lot of models I just choose Alpaca, Guanaco or ChatLm and results pretty good. lol I still don't know how to setup a custom template in Ooba. But this may help me.

RickyPossum

Jan 31, 2024

did it?

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