load arrays with up to 1024 elements when estimating

This mirrors the old behavior before #10382

load arrays with up to 1024 elements when estimating
This mirrors the old behavior before #10382
d2ee599d · Devon Rifkin · 6ed88985 · d2ee599d
Commit d2ee599d authored Apr 27, 2025 by Devon Rifkin
Hide whitespace changes
Inline Side-by-side

Showing with 1 addition and 1 deletion

server/sched.go server/sched.go +1 -1

No files found.
--- a/server/sched.go
+++ b/server/sched.go
@@ -201,7 +201,7 @@ func (s *Scheduler) processPending(ctx context.Context) {
 					}

 					// Load model for fitting
-					ggml, err := llm.LoadModel(pending.model.ModelPath, 0)
+					ggml, err := llm.LoadModel(pending.model.ModelPath, 1024)
 					if err != nil {
 						pending.errCh <- err
 						break