Skip to content

Get up and running with Kimi, GLM, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.

load arrays with up to 1024 elements when estimating

Devon Rifkind2ee599dcf97
diff --git a/server/sched.go b/server/sched.go
index d5b19fbfd..9c13f6cf5 100644
--- 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