wangzhaode
released this
19 Apr 05:27
·
59 commits
to master
since this release
Llama-3-8B-Instruct
导出onnx转换得到的int4量化版本mnn模型。
模型列表:
- tokenizer.txt
- embeddings_bf16.bin
- lm.mnn
- block_[0-31].mnn