How to use TIGER-Lab/AceCodeRM-32B with Transformers:
# Load model directly from transformers import AutoTokenizer, Qwen2ForCausalRM tokenizer = AutoTokenizer.from_pretrained("TIGER-Lab/AceCodeRM-32B") model = Qwen2ForCausalRM.from_pretrained("TIGER-Lab/AceCodeRM-32B")
Any plans to support a Qwen2ForCausalRM architecture contribution to Llama.cpp? I would really like to try this model out, given the space requirements its not feasible without quantization. Appreciate your input and thank you for reading.
· Sign up or log in to comment
