Instructions to use keyvan-ai/Mankei-Boarisch with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use keyvan-ai/Mankei-Boarisch with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="keyvan-ai/Mankei-Boarisch")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("keyvan-ai/Mankei-Boarisch")
model = AutoModelForCausalLM.from_pretrained("keyvan-ai/Mankei-Boarisch", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use keyvan-ai/Mankei-Boarisch 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 keyvan-ai/Mankei-Boarisch:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf keyvan-ai/Mankei-Boarisch:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf keyvan-ai/Mankei-Boarisch:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf keyvan-ai/Mankei-Boarisch: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 keyvan-ai/Mankei-Boarisch:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf keyvan-ai/Mankei-Boarisch: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 keyvan-ai/Mankei-Boarisch:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf keyvan-ai/Mankei-Boarisch:Q4_K_M

Use Docker

docker model run hf.co/keyvan-ai/Mankei-Boarisch:Q4_K_M

LM Studio
Jan

vLLM

How to use keyvan-ai/Mankei-Boarisch with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "keyvan-ai/Mankei-Boarisch"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "keyvan-ai/Mankei-Boarisch",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/keyvan-ai/Mankei-Boarisch:Q4_K_M

SGLang

How to use keyvan-ai/Mankei-Boarisch with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "keyvan-ai/Mankei-Boarisch" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "keyvan-ai/Mankei-Boarisch",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "keyvan-ai/Mankei-Boarisch" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "keyvan-ai/Mankei-Boarisch",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use keyvan-ai/Mankei-Boarisch with Ollama:
```
ollama run hf.co/keyvan-ai/Mankei-Boarisch:Q4_K_M
```

Unsloth Studio

How to use keyvan-ai/Mankei-Boarisch 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 keyvan-ai/Mankei-Boarisch 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 keyvan-ai/Mankei-Boarisch to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for keyvan-ai/Mankei-Boarisch to start chatting

Docker Model Runner
How to use keyvan-ai/Mankei-Boarisch with Docker Model Runner:
```
docker model run hf.co/keyvan-ai/Mankei-Boarisch:Q4_K_M
```

Lemonade

How to use keyvan-ai/Mankei-Boarisch with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull keyvan-ai/Mankei-Boarisch:Q4_K_M

Run and chat with the model

lemonade run user.Mankei-Boarisch-Q4_K_M

List all available models

lemonade list

Atomic Chat

Mankei-Boarisch

Bairisch sprechendes Chat-Sprachmodell mit rund einer Milliarde Parametern, abgeleitet aus dem deutschen Mankei-Basismodell und auf bairischen Dialekt weitertrainiert. Es führt Konversationen auf Boarisch, überträgt hochdeutschen Text in die Mundart und läuft vollständig on-premise. Apache 2.0.

Fähigkeiten

Konversation auf Boarisch — Alltag, Auskunft, Ratschlag
Übertragung hochdeutscher Texte in die bairische Mundart
Lokaler Betrieb ohne Cloud; die Daten verlassen das Haus nicht

Läuft mit

Ollama / llama.cpp — GGUF, CPU-fähig:

ollama run hf.co/keyvan-ai/Mankei-Boarisch:Q4_K_M

vLLM — Transformers-Gewichte, GPU:

vllm serve keyvan-ai/Mankei-Boarisch

GGUF-Quantisierungen

Quantisierung	Größe	Datei
f16	2,0 GB	`Mankei-Boarisch-f16.gguf`
Q8_0	1,1 GB	`Mankei-Boarisch-q8_0.gguf`
Q6_K	854 MB	`Mankei-Boarisch-q6_k.gguf`
Q5_K_M	728 MB	`Mankei-Boarisch-q5_k_m.gguf`
Q4_K_M	637 MB	`Mankei-Boarisch-q4_k_m.gguf`

Verwendung (Transformers)

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

tok = AutoTokenizer.from_pretrained("keyvan-ai/Mankei-Boarisch")
m   = AutoModelForCausalLM.from_pretrained("keyvan-ai/Mankei-Boarisch", dtype=torch.bfloat16).eval()

msgs = [{"role": "user", "content": "Servus, wia geht's da heid?"}]
enc = tok.apply_chat_template(msgs, add_generation_prompt=True, return_tensors="pt", return_dict=True)
out = m.generate(**enc, max_new_tokens=120, do_sample=True,
                 temperature=0.4, top_p=0.9, min_p=0.05, repetition_penalty=1.1)
print(tok.decode(out[0][enc["input_ids"].shape[1]:], skip_special_tokens=True))

Empfohlenes Sampling: temperature 0.4 · min_p 0.05 · repetition_penalty 1.1.

Einsatzbereich

Ausgelegt auf bairische Konversation und Mundart. Für faktische Recherche und tagesaktuelle Auskünfte sind größere Modelle und RAG-Aufbauten die geeignete Wahl.

Training

Continued Pretraining auf bairischem Wikipedia-Text (bar.wikipedia, CC-BY-SA), anschließend Supervised Fine-Tuning auf bairischen Dialogen, einem Persona-Set und Sicherheitsdaten. Basismodell: deutsches Mankei-1B. Trainiert auf offen lizenzierten und selbst erzeugten Daten.

Downloads last month: 398

Safetensors

Model size

1B params

Tensor type

BF16

keyvan-ai
/

Mankei-Boarisch