How to use Ashmal/MBZUAI-oryx with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="Ashmal/MBZUAI-oryx") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)
# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("Ashmal/MBZUAI-oryx") model = AutoModelForCausalLM.from_pretrained("Ashmal/MBZUAI-oryx") 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]:]))
How to use Ashmal/MBZUAI-oryx with vLLM:
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Ashmal/MBZUAI-oryx" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Ashmal/MBZUAI-oryx", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'
docker model run hf.co/Ashmal/MBZUAI-oryx
How to use Ashmal/MBZUAI-oryx with SGLang:
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "Ashmal/MBZUAI-oryx" \ --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": "Ashmal/MBZUAI-oryx", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'
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 "Ashmal/MBZUAI-oryx" \ --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": "Ashmal/MBZUAI-oryx", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'
How to use Ashmal/MBZUAI-oryx with Docker Model Runner:
This is the Arabic test model built at MBZUAI. More details of the projects will be announced later along with the release. This model card is just to test the capabilities of this model on Arabic benchmarks.
Chat template
Files info
