MCPcopy Create free account
hub / github.com/PaddlePaddle/FastDeploy / run_to_completion

Function run_to_completion

fastdeploy/benchmarks/latency.py:102–107  ·  view source on GitHub ↗
()

Source from the content-addressed store, hash-verified

100 llm.generate(dummy_prompts, sampling_params=sampling_params, use_tqdm=False, stream=True)
101
102 def run_to_completion():
103 start_time = time.perf_counter()
104 llm_generate()
105 end_time = time.perf_counter()
106 latency = end_time - start_time
107 return latency
108
109 print("Warming up...")
110 for _ in tqdm(range(args.num_iters_warmup), desc="Warmup iterations"):

Callers 1

mainFunction · 0.85

Calls 1

llm_generateFunction · 0.85

Tested by

no test coverage detected