This commit is contained in:
Michael Yang
2024-08-01 14:52:15 -07:00
parent 558a54b098
commit b732beba6a
68 changed files with 199 additions and 149 deletions

View File

@@ -58,7 +58,7 @@ var defaultModelsPerGPU = 3
// we'll back off down to 1 to try to get it to fit
var defaultParallel = 4
var ErrMaxQueue = fmt.Errorf("server busy, please try again. maximum pending requests exceeded")
var ErrMaxQueue = errors.New("server busy, please try again. maximum pending requests exceeded")
func InitScheduler(ctx context.Context) *Scheduler {
maxQueue := envconfig.MaxQueue()