diff --git a/llm/server.go b/llm/server.go
index e7c1029e83..079ecb2575 100644
--- a/llm/server.go
+++ b/llm/server.go
@@ -1486,7 +1486,10 @@ func (s *llmServer) Completion(ctx context.Context, req CompletionRequest, fn fu
 	serverReq.Header.Set("Content-Type", "application/json")
 
 	res, err := http.DefaultClient.Do(serverReq)
-	if err != nil {
+	if err != nil && errors.Is(err, context.Canceled) {
+		// client closed connection
+		return err
+	} else if err != nil {
 		slog.Error("post predict", "error", err)
 		return errors.New("model runner has unexpectedly stopped, this may be due to resource limitations or an internal error, check ollama server logs for details")
 	}
diff --git a/server/sched.go b/server/sched.go
index 5f8c61e099..1996ebfd85 100644
--- a/server/sched.go
+++ b/server/sched.go
@@ -229,8 +229,9 @@ func (s *Scheduler) processPending(ctx context.Context) {
 				}
 
 				if runnerToExpire == nil {
-					// Shouildn't happen
-					slog.Error("runner to expire was nil!")
+					// While we were performing load calculations, the loaded runner(s) unloaded in parallel
+					// so findRunnerToUnload returned no runners.  We'll try again and the loadedCount should be zero
+					slog.Debug("runner to expire was nil, retrying")
 					continue
 				}
 				// Trigger an expiration to unload once it's done