Unverified50% confidenceFactExact time
常见的推理引擎包括NVIDIA的TensorRT、微软的ONNX Runtime、Intel的OpenVINO
1
Sources
50%
Confidence
Long-term
Relevance
8/18/2026
First Seen
Sources
Related Claims
UnverifiedONNX Runtime是微软开源的高性能推理引擎，支持CPU、GPU、NPU等多种硬件后端，并内置图优化、算子融合等加速手段75% similarUnverifiedCPU、GPU、NPU的协同调度由英特尔的OpenVINO框架统一管理68% similarUnverifiedTensorFlow Lite、ONNX Runtime等推理框架可与底层驱动和硬件加速器（NPU/DSP）协同优化66% similarUnverifiedOllama通过ollama pull命令即可下载并运行Llama、Mistral、Qwen等主流开源模型，其推理引擎会自动检测硬件并优先调用GPU加速66% similarUnverifiedNVIDIA推理软件栈从TensorRT到TensorRT-LLM再到Triton Inference Server，构建覆盖单算子优化、单机多卡部署到大规模分布式服务的完整体系65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/771241API
curl https://kongchang.com/api/v1/knowledge/claims/771241MCP
get_claim(id=771241)