Instructions to use Blizado/German_Mistral7B_merge_tests with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Blizado/German_Mistral7B_merge_tests with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="Blizado/German_Mistral7B_merge_tests",
	filename="german_200424-q8_0.gguf",
)

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

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use Blizado/German_Mistral7B_merge_tests with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Blizado/German_Mistral7B_merge_tests:Q8_0
# Run inference directly in the terminal:
llama-cli -hf Blizado/German_Mistral7B_merge_tests:Q8_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Blizado/German_Mistral7B_merge_tests:Q8_0
# Run inference directly in the terminal:
llama-cli -hf Blizado/German_Mistral7B_merge_tests:Q8_0

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 Blizado/German_Mistral7B_merge_tests:Q8_0
# Run inference directly in the terminal:
./llama-cli -hf Blizado/German_Mistral7B_merge_tests:Q8_0

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 Blizado/German_Mistral7B_merge_tests:Q8_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Blizado/German_Mistral7B_merge_tests:Q8_0

Use Docker

docker model run hf.co/Blizado/German_Mistral7B_merge_tests:Q8_0

LM Studio
Jan

vLLM

How to use Blizado/German_Mistral7B_merge_tests with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Blizado/German_Mistral7B_merge_tests"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Blizado/German_Mistral7B_merge_tests",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Blizado/German_Mistral7B_merge_tests:Q8_0

Ollama
How to use Blizado/German_Mistral7B_merge_tests with Ollama:
```
ollama run hf.co/Blizado/German_Mistral7B_merge_tests:Q8_0
```

Unsloth Studio new

How to use Blizado/German_Mistral7B_merge_tests 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 Blizado/German_Mistral7B_merge_tests 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 Blizado/German_Mistral7B_merge_tests to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Blizado/German_Mistral7B_merge_tests to start chatting

Docker Model Runner
How to use Blizado/German_Mistral7B_merge_tests with Docker Model Runner:
```
docker model run hf.co/Blizado/German_Mistral7B_merge_tests:Q8_0
```

Lemonade

How to use Blizado/German_Mistral7B_merge_tests with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Blizado/German_Mistral7B_merge_tests:Q8_0

Run and chat with the model

lemonade run user.German_Mistral7B_merge_tests-Q8_0

List all available models

lemonade list

7B Mistral based Mergekit tests for german text generation to find the best working (for context, grammar and general output) German merge for primarily fun Chat (E)RP, but could also work for other tasks. ChatML format works well, Alpaca should work even better.

Tester feedback welcome.

The best models will be released as 16bit safetensors, all kind of GGUF quants and all needed informations.

Downloads last month: 39

GGUF

Model size

7B params

Architecture

llama

Hardware compatibility

8-bit