Instructions to use microsoft/Orca-2-13b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use microsoft/Orca-2-13b with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="microsoft/Orca-2-13b")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("microsoft/Orca-2-13b") model = AutoModelForCausalLM.from_pretrained("microsoft/Orca-2-13b") - Inference
- Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use microsoft/Orca-2-13b with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "microsoft/Orca-2-13b" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "microsoft/Orca-2-13b", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/microsoft/Orca-2-13b
- SGLang
How to use microsoft/Orca-2-13b 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 "microsoft/Orca-2-13b" \ --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": "microsoft/Orca-2-13b", "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 "microsoft/Orca-2-13b" \ --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": "microsoft/Orca-2-13b", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use microsoft/Orca-2-13b with Docker Model Runner:
docker model run hf.co/microsoft/Orca-2-13b
Repeatability
I noticed Orca2 often gives different answers if you ask the same question multiple times. For example, here is the multiple choice problem shown in the original paper. The correct answer is C. If I ask Orca2 (13B, Q4 Q8), it alternates between answer C and D. The temperature in the setting is zero. Any thoughts?
Scientists have studied the productivity of crops in mountain valleys. In some areas, the valleys are more productive than others. The increase in which factor most likely accounts for the high productivity of some areas in mountain valleys? Options:
(A)leaching of soils
(B)evaporation rates
(C)runoff from rains
(D)amounts of sunlight.