From 6c9237d09397a436f2d9218fc1e1e62c83d2f8e0 Mon Sep 17 00:00:00 2001
From: DefTruth <31974251+DefTruth@users.noreply.github.com>
Date: Thu, 26 Sep 2024 00:35:53 +0800
Subject: [PATCH] [VLM][Bugfix] internvl with num_scheduler_steps > 1 (#8614)

---
 vllm/model_executor/models/internvl.py | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/vllm/model_executor/models/internvl.py b/vllm/model_executor/models/internvl.py
index fffd0d4161e10..b1748700d481a 100644
--- a/vllm/model_executor/models/internvl.py
+++ b/vllm/model_executor/models/internvl.py
@@ -19,7 +19,7 @@
 from vllm.distributed import get_pp_group
 from vllm.inputs import INPUT_REGISTRY, InputContext, LLMInputs
 from vllm.model_executor.layers.quantization import QuantizationConfig
-from vllm.model_executor.layers.sampler import SamplerOutput
+from vllm.model_executor.layers.sampler import Sampler, SamplerOutput
 from vllm.model_executor.model_loader.weight_utils import default_weight_loader
 from vllm.model_executor.models.intern_vit import InternVisionModel
 from vllm.model_executor.sampling_metadata import SamplingMetadata
@@ -379,6 +379,11 @@ def __init__(self,
         self.make_empty_intermediate_tensors = (
             self.language_model.make_empty_intermediate_tensors)
 
+        if hasattr(self.language_model, "sampler"):
+            self.sampler = self.language_model.sampler
+        else:
+            self.sampler = Sampler()
+
     def pixel_shuffle(self, x, scale_factor=0.5):
         n, w, h, c = x.size()
         # N, W, H, C --> N, W, H * scale, C // scale