How to use OpenVINO/Qwen3-Reranker-0.6B-seq-cls-fp16-ov with Transformers:
# Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("OpenVINO/Qwen3-Reranker-0.6B-seq-cls-fp16-ov") model = AutoModelForSequenceClassification.from_pretrained("OpenVINO/Qwen3-Reranker-0.6B-seq-cls-fp16-ov", device_map="auto")
NPU加速用不了呀？全程igpu干活，时npu不支持embedding层结构么？
· Sign up or log in to comment
