Files
LocalAI/core/backend/video.go
Ettore Di Giacinto fc5b9ebfcc feat(loader): enhance single active backend to support LRU eviction (#7535)
* feat(loader): refactor single active backend support to LRU

This changeset introduces LRU management of loaded backends. Users can
set now a maximum number of models to be loaded concurrently, and, when
setting LocalAI in single active backend mode we set LRU to 1 for
backward compatibility.

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

* chore: add tests

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

* Update docs

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

* Fixups

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

---------

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
2025-12-12 12:28:38 +01:00

42 lines
1.1 KiB
Go

package backend
import (
"github.com/mudler/LocalAI/core/config"
"github.com/mudler/LocalAI/pkg/grpc/proto"
model "github.com/mudler/LocalAI/pkg/model"
)
func VideoGeneration(height, width int32, prompt, negativePrompt, startImage, endImage, dst string, numFrames, fps, seed int32, cfgScale float32, step int32, loader *model.ModelLoader, modelConfig config.ModelConfig, appConfig *config.ApplicationConfig) (func() error, error) {
opts := ModelOptions(modelConfig, appConfig)
inferenceModel, err := loader.Load(
opts...,
)
if err != nil {
return nil, err
}
fn := func() error {
_, err := inferenceModel.GenerateVideo(
appConfig.Context,
&proto.GenerateVideoRequest{
Height: height,
Width: width,
Prompt: prompt,
NegativePrompt: negativePrompt,
StartImage: startImage,
EndImage: endImage,
NumFrames: numFrames,
Fps: fps,
Seed: seed,
CfgScale: cfgScale,
Step: step,
Dst: dst,
})
return err
}
return fn, nil
}