diff --git a/mteb/models/openai_models.py b/mteb/models/openai_models.py
index adf96fbe4e..aecacf549a 100644
--- a/mteb/models/openai_models.py
+++ b/mteb/models/openai_models.py
@@ -63,7 +63,7 @@ def encode(self, sentences: list[str], **kwargs: Any) -> np.ndarray:
             else:
                 trimmed_sentences.append(sentence)
 
-        max_batch_size = 2048
+        max_batch_size = kwargs.get("batch_size", 2048)
         sublists = [
             trimmed_sentences[i : i + max_batch_size]
             for i in range(0, len(trimmed_sentences), max_batch_size)