From: Alexey Parfenov <redacted>
Date: Sat, 28 Dec 2024 15:08:54 +0000 (+0000)
Subject: server : fix token duplication when streaming with stop strings (#10997)
X-Git-Tag: upstream/0.0.4488~94
X-Git-Url: https://git.djapps.eu/?a=commitdiff_plain;h=16cdce7b68218959e0658e2f95b4572573d5008e;p=pkg%2Fggml%2Fsources%2Fllama.cpp

server : fix token duplication when streaming with stop strings (#10997)
---

diff --git a/examples/server/server.cpp b/examples/server/server.cpp
index 30ff3b14..3558ddb7 100644
--- a/examples/server/server.cpp
+++ b/examples/server/server.cpp
@@ -1856,6 +1856,8 @@ struct server_context {
                 result.text_to_send = slot.generated_text.substr(pos, std::string::npos);
                 slot.n_sent_text += result.text_to_send.size();
                 // add the token to slot queue and cache
+            } else {
+                result.text_to_send = "";
             }
 
             slot.add_token(result);