This commit is contained in:
Josh Yan 2024-07-09 10:27:33 -07:00
parent e87eafe5cd
commit 1344843515

View File

@ -440,6 +440,8 @@ func CreateModel(ctx context.Context, name model.Name, modelFileDir, quantizatio
defer temp.Close() defer temp.Close()
defer os.Remove(temp.Name()) defer os.Remove(temp.Name())
// Quantizes per layer
// Save total quantized tensors
if err := llm.Quantize(blob, temp.Name(), want); err != nil { if err := llm.Quantize(blob, temp.Name(), want); err != nil {
return err return err
} }
@ -480,7 +482,7 @@ func CreateModel(ctx context.Context, name model.Name, modelFileDir, quantizatio
} }
fn(api.ProgressResponse{ fn(api.ProgressResponse{
Status: fmt.Sprintf("quantizing model %d%%", 100), Status: "quantizing model done",
Quantize: quantization, Quantize: quantization,
}) })
case "license", "template", "system": case "license", "template", "system":