load arrays with up to 1024 elements when estimating

This mirrors the old behavior before #10382
This commit is contained in:
Devon Rifkin 2025-04-27 13:45:13 -07:00
parent 6ed8898590
commit d2ee599dcf

View File

@ -201,7 +201,7 @@ func (s *Scheduler) processPending(ctx context.Context) {
} }
// Load model for fitting // Load model for fitting
ggml, err := llm.LoadModel(pending.model.ModelPath, 0) ggml, err := llm.LoadModel(pending.model.ModelPath, 1024)
if err != nil { if err != nil {
pending.errCh <- err pending.errCh <- err
break break