Instructions to use lightblue/Karasu-Mixtral-8x22B-v0.1-gguf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use lightblue/Karasu-Mixtral-8x22B-v0.1-gguf with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="lightblue/Karasu-Mixtral-8x22B-v0.1-gguf",
	filename="Karasu-Mixtral-8x22B-v0.1-Q3_K_M-00001-of-00005.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use lightblue/Karasu-Mixtral-8x22B-v0.1-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 lightblue/Karasu-Mixtral-8x22B-v0.1-gguf:Q3_K_M
# Run inference directly in the terminal:
llama cli -hf lightblue/Karasu-Mixtral-8x22B-v0.1-gguf:Q3_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf lightblue/Karasu-Mixtral-8x22B-v0.1-gguf:Q3_K_M
# Run inference directly in the terminal:
llama cli -hf lightblue/Karasu-Mixtral-8x22B-v0.1-gguf:Q3_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 lightblue/Karasu-Mixtral-8x22B-v0.1-gguf:Q3_K_M
# Run inference directly in the terminal:
./llama-cli -hf lightblue/Karasu-Mixtral-8x22B-v0.1-gguf:Q3_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 lightblue/Karasu-Mixtral-8x22B-v0.1-gguf:Q3_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf lightblue/Karasu-Mixtral-8x22B-v0.1-gguf:Q3_K_M

Use Docker

docker model run hf.co/lightblue/Karasu-Mixtral-8x22B-v0.1-gguf:Q3_K_M

LM Studio
Jan
Ollama
How to use lightblue/Karasu-Mixtral-8x22B-v0.1-gguf with Ollama:
```
ollama run hf.co/lightblue/Karasu-Mixtral-8x22B-v0.1-gguf:Q3_K_M
```

Unsloth Studio

How to use lightblue/Karasu-Mixtral-8x22B-v0.1-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 lightblue/Karasu-Mixtral-8x22B-v0.1-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 lightblue/Karasu-Mixtral-8x22B-v0.1-gguf to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for lightblue/Karasu-Mixtral-8x22B-v0.1-gguf to start chatting

Atomic Chat new
Docker Model Runner
How to use lightblue/Karasu-Mixtral-8x22B-v0.1-gguf with Docker Model Runner:
```
docker model run hf.co/lightblue/Karasu-Mixtral-8x22B-v0.1-gguf:Q3_K_M
```

Lemonade

How to use lightblue/Karasu-Mixtral-8x22B-v0.1-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull lightblue/Karasu-Mixtral-8x22B-v0.1-gguf:Q3_K_M

Run and chat with the model

lemonade run user.Karasu-Mixtral-8x22B-v0.1-gguf-Q3_K_M

List all available models

lemonade list

YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

This is the Q3_K_M GGUF port of the lightblue/Karasu-Mixtral-8x22B-v0.1 model.

How to use

The way to run this directly are using the llama.cpp package.

git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp
make
huggingface-cli download lightblue/Karasu-Mixtral-8x22B-v0.1-gguf  --local-dir /some/folder/
./main -m /some/folder/Karasu-Mixtral-8x22B-v0.1-Q3_K_M-00001-of-00005.gguf -p "<s>[INST] Tell me a really funny joke. No puns! [/INST]" -n 256 -e

If you would like a nice easy GUI and have >64GB of RAM, then you could also run this using LM Studio and search for this model on the search bar.

Commands to make this:

cd llama.cpp

./convert-hf-to-gguf.py --outfile  /workspace/Karasu-Mixtral-8x22B-v0.1.gguf --outtype f16 /workspace/llm_training/axolotl/mixtral_8x22B_training/merged_model_multiling

./quantize /workspace/Karasu-Mixtral-8x22B-v0.1.gguf /workspace/Karasu-Mixtral-8x22B-v0.1-Q3_K_M.gguf Q3_K_M

./gguf-split  --split --split-max-tensors 128  /workspace/Karasu-Mixtral-8x22B-v0.1-Q3_K_M.gguf /workspace/split_gguf_q3km/Karasu-Mixtral-8x22B-v0.1-Q3_K_M

Downloads last month: 9

GGUF

Model size

141B params

Architecture

llama

Hardware compatibility

3-bit

Inference Providers NEW

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