Instructions to use Mwanzau/Mazgu_Llama_V3-Merged with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Mwanzau/Mazgu_Llama_V3-Merged with Transformers:

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

pipe = pipeline("text-generation", model="Mwanzau/Mazgu_Llama_V3-Merged")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Mwanzau/Mazgu_Llama_V3-Merged")
model = AutoModelForCausalLM.from_pretrained("Mwanzau/Mazgu_Llama_V3-Merged", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Mwanzau/Mazgu_Llama_V3-Merged with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Mwanzau/Mazgu_Llama_V3-Merged"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Mwanzau/Mazgu_Llama_V3-Merged",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Mwanzau/Mazgu_Llama_V3-Merged

SGLang

How to use Mwanzau/Mazgu_Llama_V3-Merged 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 "Mwanzau/Mazgu_Llama_V3-Merged" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Mwanzau/Mazgu_Llama_V3-Merged",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "Mwanzau/Mazgu_Llama_V3-Merged" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Mwanzau/Mazgu_Llama_V3-Merged",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Unsloth Studio

How to use Mwanzau/Mazgu_Llama_V3-Merged 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 Mwanzau/Mazgu_Llama_V3-Merged 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 Mwanzau/Mazgu_Llama_V3-Merged to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Mwanzau/Mazgu_Llama_V3-Merged to start chatting

Load model with FastModel

pip install unsloth
from unsloth import FastModel
model, tokenizer = FastModel.from_pretrained(
    model_name="Mwanzau/Mazgu_Llama_V3-Merged",
    max_seq_length=2048,
)

Docker Model Runner
How to use Mwanzau/Mazgu_Llama_V3-Merged with Docker Model Runner:
```
docker model run hf.co/Mwanzau/Mazgu_Llama_V3-Merged
```

Uploaded finetuned model

Developed by: Mwanzau
License: apache-2.0
Finetuned from model : Mwanzau/Mazgu_Llama_V3-Merged

This llama model was trained 2x faster with Unsloth and Huggingface's TRL library.

current output samples

--- Testing Prompt: 'Bantu ba mu Malawi bakutemwa' --- <|begin_of_text|>Bantu ba mu Malawi bakutemwa ghakuti munthu ŵakuwuka kuti vinyake viwo ya mwaka waliyose kuŴa chikuzizilira. Ndipo kukhumba ma vinji yakujomazgika cha ndimu yayo. Nyengo yachitiri

Pa mahana na mwechita, nyumbanga likamwanapo pa nkhani nandi za vyambura panyemi ku muntha gha mal

--- Testing Prompt: 'Munda wane uli na' --- <|begin_of_text|>Munda wane uli na ghakupangiska kutiwuka kwa vyose wa kukhaluma pa chimuwona ŵamanya mulembiya munthu ya nyengo yilengisuzga, kuŴa ngami Kuti nkhani zikusanga vintha mu ndipo prachoko vya mwakuŵaka yayo. Ndiposo pakughanja pali panji nkhi na ku kusungika uyo malapo makora

--- Testing Prompt: 'Chitumbuka chiri na mazgu gha' --- <|begin_of_text|>Chitumbuka chiri na mazgu gha kuziziga kuti ŵakulondeka mwa muntha

Kusanga mwali, mu vinthu ya ku nyengo wake. Nyumba yikwamba chita yayo ghafuma cha vinyowoya panji ivyo ndimu yichomene.

Nyaka 6 uwo nkhana la masoma chikuwezga avya zimanyeka mahondo. kweniya ma vyose yakufumika

Downloads last month: 301

Safetensors

Model size

1B params

Tensor type

BF16

Model tree for Mwanzau/Mazgu_Llama_V3-Merged

Unable to build the model tree, the base model loops to the model itself. Learn more.