dearwolf/LocalAI
1
0
Fork 0
mirror of https://github.com/mudler/LocalAI.git synced 2025-05-20 10:35:01 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 21

feat: add --single-active-backend to allow only one backend active at the time (#925)

Browse source 
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
This commit is contained in:
Ettore Di Giacinto 2023-08-19 01:49:33 +02:00 • committed by GitHub
parent 1079b18ff7
commit afdc0ebfd7
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
20 changed files with 238 additions and 164 deletions
Download patch file Download diff file Expand all files Collapse all files

16
api/backend/llm.go
View file

@ -33,25 +33,13 @@ func ModelInference(ctx context.Context, s string, loader *model.ModelLoader, c
var inferenceModel *grpc.Client
var err error
opts := []model.Option{
opts := modelOpts(c, o, []model.Option{
model.WithLoadGRPCLoadModelOpts(grpcOpts),
model.WithThreads(uint32(c.Threads)), // some models uses this to allocate threads during startup
model.WithAssetDir(o.AssetsDestination),
model.WithModel(modelFile),
model.WithContext(o.Context),
}
if c.GRPC.Attempts != 0 {
opts = append(opts, model.WithGRPCAttempts(c.GRPC.Attempts))
}
if c.GRPC.AttemptsSleepTime != 0 {
opts = append(opts, model.WithGRPCAttemptsDelay(c.GRPC.AttemptsSleepTime))
}
for k, v := range o.ExternalGRPCBackends {
opts = append(opts, model.WithExternalBackend(k, v))
}
})
if c.Backend != "" {
opts = append(opts, model.WithBackendString(c.Backend))