From 6ed540b52f091b8e638e2f8dbe911238310dfc5f Mon Sep 17 00:00:00 2001
From: Mohit Sharma <mohit21sharma.ms@gmail.com>
Date: Mon, 21 Apr 2025 15:28:18 +0000
Subject: [PATCH] add improvements

---
 server/text_generation_server/models/flash_causal_lm.py | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/server/text_generation_server/models/flash_causal_lm.py b/server/text_generation_server/models/flash_causal_lm.py
index 9c1d2225..581cbde8 100644
--- a/server/text_generation_server/models/flash_causal_lm.py
+++ b/server/text_generation_server/models/flash_causal_lm.py
@@ -1346,6 +1346,9 @@ class FlashCausalLM(Model):
     def batch_type(self) -> Type[FlashCausalLMBatch]:
         return FlashCausalLMBatch
 
+    def get_input_embeddings(self, batch):
+        batch.inputs_embeds = None
+
     def init_kv_cache(
         self,
         num_blocks: int,