From 8c508fa8fbe29a3c3e1b3d2ae3cea7ab37e9c41c Mon Sep 17 00:00:00 2001 From: Alvaro Moran Date: Mon, 9 Sep 2024 13:10:37 +0000 Subject: [PATCH] chore(warmup): removed superflous prefill call --- .../text_generation_server/jetstream_pt_support/generator.py | 2 -- 1 file changed, 2 deletions(-) diff --git a/text-generation-inference/server/text_generation_server/jetstream_pt_support/generator.py b/text-generation-inference/server/text_generation_server/jetstream_pt_support/generator.py index b3bc00a..caa412d 100644 --- a/text-generation-inference/server/text_generation_server/jetstream_pt_support/generator.py +++ b/text-generation-inference/server/text_generation_server/jetstream_pt_support/generator.py @@ -351,8 +351,6 @@ def warmup(self, batch: Batch) -> int: if not decode_done: logger.debug("No decode done during warmup") - self.prefill(batch) - self.clear() elapsed = time.time() - start logger.debug(f"Warmup done, took {elapsed:.2f}s") seq_len = self.engine.env.seq_len