Instructions to use anthracite-org/magnum-v2.5-12b-kto-gguf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use anthracite-org/magnum-v2.5-12b-kto-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 anthracite-org/magnum-v2.5-12b-kto-gguf:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf anthracite-org/magnum-v2.5-12b-kto-gguf:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf anthracite-org/magnum-v2.5-12b-kto-gguf:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf anthracite-org/magnum-v2.5-12b-kto-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 anthracite-org/magnum-v2.5-12b-kto-gguf:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf anthracite-org/magnum-v2.5-12b-kto-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 anthracite-org/magnum-v2.5-12b-kto-gguf:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf anthracite-org/magnum-v2.5-12b-kto-gguf:Q4_K_M

Use Docker

docker model run hf.co/anthracite-org/magnum-v2.5-12b-kto-gguf:Q4_K_M

LM Studio
Jan

vLLM

How to use anthracite-org/magnum-v2.5-12b-kto-gguf with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "anthracite-org/magnum-v2.5-12b-kto-gguf"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "anthracite-org/magnum-v2.5-12b-kto-gguf",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/anthracite-org/magnum-v2.5-12b-kto-gguf:Q4_K_M

Ollama
How to use anthracite-org/magnum-v2.5-12b-kto-gguf with Ollama:
```
ollama run hf.co/anthracite-org/magnum-v2.5-12b-kto-gguf:Q4_K_M
```
Unsloth Desktop
Docker Model Runner
How to use anthracite-org/magnum-v2.5-12b-kto-gguf with Docker Model Runner:
```
docker model run hf.co/anthracite-org/magnum-v2.5-12b-kto-gguf:Q4_K_M
```

Lemonade

How to use anthracite-org/magnum-v2.5-12b-kto-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull anthracite-org/magnum-v2.5-12b-kto-gguf:Q4_K_M

Run and chat with the model

lemonade run user.magnum-v2.5-12b-kto-gguf-Q4_K_M

List all available models

lemonade list

Atomic Chat

This repo contains GGUF quants of the model. If you need the original weights, please find them here.

imatrix_data included

v2.5 KTO is an experimental release; we are testing a hybrid reinforcement learning strategy of KTO + DPOP, using rejected data sampled from the original model as "rejected". For "chosen", we use data from the original finetuning dataset as "chosen". This was done on a limited portion of of primarily instruction following data; we plan to scale up a larger KTO dataset in the future for better generalization.

This is the 5th in a series of models designed to replicate the prose quality of the Claude 3 models, specifically Sonnet and Opus. This model is fine-tuned on top of anthracite-org/magnum-12b-v2.

Prompting

Model has been Instruct tuned with the ChatML formatting. A typical input would look like this:

"""<|im_start|>user
Hi there!<|im_end|>
<|im_start|>assistant
Nice to meet you!<|im_end|>
<|im_start|>user
Can I ask a question?<|im_end|>
<|im_start|>assistant
"""