From 731a345cfc0ba503d05a95630dc86e6a01d35b70 Mon Sep 17 00:00:00 2001
From: DocShotgun <126566557+DocShotgun@users.noreply.github.com>
Date: Tue, 19 Nov 2024 12:40:00 -0800
Subject: [PATCH] OAI: Keep behavior consistent between chat completion and
 encode * When vision is not enabled, only the first text block is kept in
 message.content if it is a list

---
 endpoints/core/router.py | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/endpoints/core/router.py b/endpoints/core/router.py
index 0a48a2e..d7837f8 100644
--- a/endpoints/core/router.py
+++ b/endpoints/core/router.py
@@ -379,6 +379,20 @@ async def encode_tokens(data: TokenEncodeRequest) -> TokenEncodeResponse:
         if model.container.use_vision:
             data.text, embeddings = await preprocess_vision_request(data.text)
 
+        # Keeping behavior consistent with format_prompt_with_template
+        # Deal with list in messages.content
+        # Just replace the content list with the very first text message
+        for message in data.text:
+            if isinstance(message["content"], list):
+                message["content"] = next(
+                    (
+                        content["text"]
+                        for content in message["content"]
+                        if content["type"] == "text"
+                    ),
+                    "",
+                )
+
         special_tokens_dict = model.container.get_special_tokens(
             unwrap(data.add_bos_token, True)
         )