Open Issues Need Help
View All on GitHubAI Summary: The LLM-TPU project requires a new C++ inference demonstration for the Phi-4 model. This demo should replicate the functionality and output quality of the existing Python demo, with the Qwen3 C++ example provided as a reference for implementation.
Machine learning compiler based on MLIR for Sophgo TPU.
AI Summary: The task is to implement a C++ inference demo for the Llama3.2 Vision model within the LLM-TPU project. This C++ demo should replicate the functionality and output quality of the existing Python demo, with a reference C++ example from Qwen2.5VL available for guidance.
Machine learning compiler based on MLIR for Sophgo TPU.
AI Summary: This issue requests the implementation of a C++ inference demo for the MiniCPM4 model within the LLM-TPU project. The goal is to replicate the functionality and output quality of the existing Python demo, with a reference C++ example from Qwen3 provided for guidance.
Machine learning compiler based on MLIR for Sophgo TPU.
Machine learning compiler based on MLIR for Sophgo TPU.
Machine learning compiler based on MLIR for Sophgo TPU.
Machine learning compiler based on MLIR for Sophgo TPU.
AI Summary: The issue requests the implementation of a C++ inference demo for the GLM-4.1V model within the LLM-TPU project. This new C++ demo should replicate the functionality and output quality of the existing Python version, with a reference example provided in Qwen2.5VL to guide the development.
Machine learning compiler based on MLIR for Sophgo TPU.
Machine learning compiler based on MLIR for Sophgo TPU.