yang.tian
yang.tian
### Git commit dc39a5e7a84815a90fa0c515ed8927870cf858c9 ### Operating systems Linux ### GGML backends Vulkan ### Problem description & steps to reproduce When I try to cross compile llama.cpp with vulkan backend support...
使用最新的代码,先安装依赖 然后python3 llmexport.py --path Qwen3-8B --export mnn --dst_path Qwen3-8B-mnn导出模型的时候有如下的报错 💥 Failed export onnx model to Traceback (most recent call last): File "/data/sw-build/andy.tian/miniforge3/envs/mnn_llm_quant/lib/python3.12/site-packages/torch/onnx/_internal/exporter/_capture_strategies.py", line 118, in __call__ exported_program = self._capture(model, args,...
commit f699ca2842cf9752d0cd90a4a5b3664c8e893ba8 Author: yanzhang Date: Fri Aug 1 17:19:36 2025 +0800 Replace the KleidiAI macro with the hint config Signed-off-by: yanzhang Change-Id: I9dcc4f68e1e67a11266b66589006650b197cde1e 在这个 patch 中增加了一个 kleidiai 的 hint config...