Instructions to use ilsp/Meltemi-7B-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use ilsp/Meltemi-7B-v1 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="ilsp/Meltemi-7B-v1", device_map="auto")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("ilsp/Meltemi-7B-v1") model = AutoModelForCausalLM.from_pretrained("ilsp/Meltemi-7B-v1", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use ilsp/Meltemi-7B-v1 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "ilsp/Meltemi-7B-v1" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "ilsp/Meltemi-7B-v1", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/ilsp/Meltemi-7B-v1
- SGLang
How to use ilsp/Meltemi-7B-v1 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 "ilsp/Meltemi-7B-v1" \ --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": "ilsp/Meltemi-7B-v1", "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 "ilsp/Meltemi-7B-v1" \ --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": "ilsp/Meltemi-7B-v1", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use ilsp/Meltemi-7B-v1 with Docker Model Runner:
docker model run hf.co/ilsp/Meltemi-7B-v1
Future plans (Llama 3?)
#3
by velocity - opened
Hello! Love this project - are there any plans on a Llama 3 version?
It seems to perform much better than Mistral and the new tokenizer should make it easier to train on Greek.
Wish I had the resources for pretraining or finetuning of this scale... As far as I can tell you're the only ones putting Greek models out so, our only hope!
Hi, thank you very much for your kind words!
We have no immediate plans for a LLAMA 3 version, but we 're considering our options.
Stay tuned for new releases soon..
velocity changed discussion status to closed