From e58174cecbc45bf79bf653cd2c984395940c6ef4 Mon Sep 17 00:00:00 2001
From: Georgi Gerganov <ggerganov@gmail.com>
Date: Thu, 18 Sep 2025 12:47:56 +0300
Subject: [PATCH] llama : bump max seq limit from 64 to 256 (#15916)

ggml-ci
---
 src/llama-cparams.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/llama-cparams.h b/src/llama-cparams.h
index dbbaba9f62..eae7b839f4 100644
--- a/src/llama-cparams.h
+++ b/src/llama-cparams.h
@@ -4,7 +4,7 @@
 
 #include <cstdint>
 
-#define LLAMA_MAX_SEQ 64
+#define LLAMA_MAX_SEQ 256
 
 struct llama_cparams {
     uint32_t n_ctx;           // context size used during inference