From c5ff443b9f7264d0973dcc2ce671d0ff174cc34f Mon Sep 17 00:00:00 2001
From: Daniel Hiltgen <daniel@ollama.com>
Date: Tue, 9 Apr 2024 16:35:10 -0700
Subject: [PATCH] Handle very slow model loads

During testing, we're seeing some models take over 3 minutes.
---
 llm/server.go | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/llm/server.go b/llm/server.go
index 0e084d5a..160effe7 100644
--- a/llm/server.go
+++ b/llm/server.go
@@ -383,7 +383,8 @@ func (s *LlamaServer) Ping(ctx context.Context) error {
 
 func (s *LlamaServer) waitUntilRunning() error {
 	start := time.Now()
-	expiresAt := time.Now().Add(3 * time.Minute) // be generous with timeout, large models can take a while to load
+	// TODO we need to wire up a better way to detect hangs during model load and startup of the server
+	expiresAt := time.Now().Add(10 * time.Minute) // be generous with timeout, large models can take a while to load
 	ticker := time.NewTicker(50 * time.Millisecond)
 	defer ticker.Stop()