From 476a066830364f868a88a5a8c2125c42731f9c67 Mon Sep 17 00:00:00 2001
From: sixsixcoder <evethwillbeok@outlook.com>
Date: Tue, 12 Nov 2024 16:45:17 +0800
Subject: [PATCH] support vllm 0.6.3
---
basic_demo/vllm_cli_vision_demo.py | 5 +----
1 file changed, 1 insertion(+), 4 deletions(-)
diff --git a/basic_demo/vllm_cli_vision_demo.py b/basic_demo/vllm_cli_vision_demo.py
index 1ed67d5..d5ad6be 100644
--- a/basic_demo/vllm_cli_vision_demo.py
+++ b/basic_demo/vllm_cli_vision_demo.py
@@ -46,9 +46,6 @@ async def vllm_gen(messages: List[Dict[str, str]], top_p: float, temperature: fl
"temperature": temperature,
"top_p": top_p,
"top_k": -1,
- "use_beam_search": False,
- "length_penalty": 1,
- "early_stopping": False,
"ignore_eos": False,
"max_tokens": max_dec_len,
"logprobs": None,
@@ -58,7 +55,7 @@ async def vllm_gen(messages: List[Dict[str, str]], top_p: float, temperature: fl
}
sampling_params = SamplingParams(**params_dict)
- async for output in engine.generate(inputs=inputs, sampling_params=sampling_params, request_id=f"{time.time()}"):
+ async for output in engine.generate(prompt=inputs, sampling_params=sampling_params, request_id=f"{time.time()}"):
yield output.outputs[0].text