How to use FastFlowLM/Hy-MT2-1.8B-NPU2 with Transformers:
# Use a pipeline as a high-level helper # Warning: Pipeline type "translation" is no longer supported in transformers v5. # You must load the model directly (see below) or downgrade to v4.x with: # pip install "transformers<5.0.0" from transformers import pipeline pipe = pipeline("translation", model="FastFlowLM/Hy-MT2-1.8B-NPU2")
# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("FastFlowLM/Hy-MT2-1.8B-NPU2") model = AutoModelForCausalLM.from_pretrained("FastFlowLM/Hy-MT2-1.8B-NPU2", device_map="auto")
https://huggingface.co/FastFlowLM/Hy-MT2-1.8B-NPU2/resolve/main/tokenizer.json
hf download hf://FastFlowLM/Hy-MT2-1.8B-NPU2/tokenizer.json
curl -L -o tokenizer.json https://huggingface.co/FastFlowLM/Hy-MT2-1.8B-NPU2/resolve/main/tokenizer.json
