§ tools · storyline
Peking University and StepFun unveil TensorCast up to 93.2% latency
Peking University and StepFun unveil TensorCast, a programmable tensor management layer that reduces LLM time-to-first-token latency by up to 93.2%.
Peking University and StepFun Unveil TensorCast: A Programmable Tensor Management Layer That Cuts LLM Time-to-First-Token by Up to 93.2% Pandaily
§ sources1 publication · timeline below