v0.1.0

Performance model update for LLMServingSim

New features

• Support GPU with a performance model
• Auto config generator (network and memory)
• Verbose option for more detailed log
• More metrics (queuing_delay, TTFT, TPOT)
• Refactored code structure for readability

v0.0.0-artifact

IISWC Artifact for "LLMServingSim: A HW/SW Co-Simulation Infrastructure for LLM Inference Serving at Scale"