From 3b5a4a77f3a191e368af3412e5de9b38b4f80771 Mon Sep 17 00:00:00 2001
From: royjhan <65097070+royjhan@users.noreply.github.com>
Date: Wed, 3 Jul 2024 13:46:23 -0700
Subject: [PATCH] Return Correct Prompt Eval Count Regardless of Cache Prompt
 (#5371)

* openai compatibility

* Revert "openai compatibility"

This reverts commit d3f98a811e00fc497d889c8c45b0cfec5b64690c.

* remove erroneous subtraction of prompt cache
---
 llm/ext_server/server.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/llm/ext_server/server.cpp b/llm/ext_server/server.cpp
index 3bc01252..09970599 100644
--- a/llm/ext_server/server.cpp
+++ b/llm/ext_server/server.cpp
@@ -1732,7 +1732,7 @@ struct llama_server_context
                             slot.n_past -= 1;
                         }
 
-                        slot.n_prompt_tokens_processed = slot.n_prompt_tokens - slot.n_past;
+                        slot.n_prompt_tokens_processed = slot.n_prompt_tokens;
 
                         if (slot.ga_n != 1)
                         {