dearwolf/LocalAI
1
0
Fork 0
mirror of https://github.com/mudler/LocalAI.git synced 2025-05-30 23:44:59 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 36

feat: Centralized Request Processing middleware (#3847)

Browse source 
* squash past, centralize request middleware PR

Signed-off-by: Dave Lee <dave@gray101.com>

* migrate bruno request files to examples repo

Signed-off-by: Dave Lee <dave@gray101.com>

* fix

Signed-off-by: Dave Lee <dave@gray101.com>

* Update tests/e2e-aio/e2e_test.go

Signed-off-by: Ettore Di Giacinto <mudler@users.noreply.github.com>

---------

Signed-off-by: Dave Lee <dave@gray101.com>
Signed-off-by: Ettore Di Giacinto <mudler@users.noreply.github.com>
Co-authored-by: Ettore Di Giacinto <mudler@users.noreply.github.com>
This commit is contained in:
Dave 2025-02-10 06:06:16 -05:00 • committed by GitHub
parent c330360785
commit 3cddf24747
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
53 changed files with 240975 additions and 821 deletions
Download patch file Download diff file Expand all files Collapse all files

13
core/http/routes/elevenlabs.go
View file

@ -4,17 +4,26 @@ import (
"github.com/gofiber/fiber/v2"
"github.com/mudler/LocalAI/core/config"
"github.com/mudler/LocalAI/core/http/endpoints/elevenlabs"
"github.com/mudler/LocalAI/core/http/middleware"
"github.com/mudler/LocalAI/core/schema"
"github.com/mudler/LocalAI/pkg/model"
)
func RegisterElevenLabsRoutes(app *fiber.App,
re *middleware.RequestExtractor,
cl *config.BackendConfigLoader,
ml *model.ModelLoader,
appConfig *config.ApplicationConfig) {
// Elevenlabs
app.Post("/v1/text-to-speech/:voice-id", elevenlabs.TTSEndpoint(cl, ml, appConfig))
app.Post("/v1/text-to-speech/:voice-id",
re.BuildFilteredFirstAvailableDefaultModel(config.BuildUsecaseFilterFn(config.FLAG_TTS)),
re.SetModelAndConfig(func() schema.LocalAIRequest { return new(schema.ElevenLabsTTSRequest) }),
elevenlabs.TTSEndpoint(cl, ml, appConfig))
app.Post("/v1/sound-generation", elevenlabs.SoundGenerationEndpoint(cl, ml, appConfig))
app.Post("/v1/sound-generation",
re.BuildFilteredFirstAvailableDefaultModel(config.BuildUsecaseFilterFn(config.FLAG_SOUND_GENERATION)),
re.SetModelAndConfig(func() schema.LocalAIRequest { return new(schema.ElevenLabsSoundGenerationRequest) }),
elevenlabs.SoundGenerationEndpoint(cl, ml, appConfig))
}

8
core/http/routes/jina.go
View file

@ -3,16 +3,22 @@ package routes
import (
"github.com/mudler/LocalAI/core/config"
"github.com/mudler/LocalAI/core/http/endpoints/jina"
"github.com/mudler/LocalAI/core/http/middleware"
"github.com/mudler/LocalAI/core/schema"
"github.com/gofiber/fiber/v2"
"github.com/mudler/LocalAI/pkg/model"
)
func RegisterJINARoutes(app *fiber.App,
re *middleware.RequestExtractor,
cl *config.BackendConfigLoader,
ml *model.ModelLoader,
appConfig *config.ApplicationConfig) {
// POST endpoint to mimic the reranking
app.Post("/v1/rerank", jina.JINARerankEndpoint(cl, ml, appConfig))
app.Post("/v1/rerank",
re.BuildFilteredFirstAvailableDefaultModel(config.BuildUsecaseFilterFn(config.FLAG_RERANK)),
re.SetModelAndConfig(func() schema.LocalAIRequest { return new(schema.JINARerankRequest) }),
jina.JINARerankEndpoint(cl, ml, appConfig))
}

28
core/http/routes/localai.go
View file

@ -5,13 +5,16 @@ import (
"github.com/gofiber/swagger"
"github.com/mudler/LocalAI/core/config"
"github.com/mudler/LocalAI/core/http/endpoints/localai"
"github.com/mudler/LocalAI/core/http/middleware"
"github.com/mudler/LocalAI/core/p2p"
"github.com/mudler/LocalAI/core/schema"
"github.com/mudler/LocalAI/core/services"
"github.com/mudler/LocalAI/internal"
"github.com/mudler/LocalAI/pkg/model"
)
func RegisterLocalAIRoutes(router *fiber.App,
requestExtractor *middleware.RequestExtractor,
cl *config.BackendConfigLoader,
ml *model.ModelLoader,
appConfig *config.ApplicationConfig,
@ -33,8 +36,18 @@ func RegisterLocalAIRoutes(router *fiber.App,
router.Get("/models/jobs", modelGalleryEndpointService.GetAllStatusEndpoint())
}
router.Post("/tts", localai.TTSEndpoint(cl, ml, appConfig))
router.Post("/vad", localai.VADEndpoint(cl, ml, appConfig))
router.Post("/tts",
requestExtractor.BuildFilteredFirstAvailableDefaultModel(config.BuildUsecaseFilterFn(config.FLAG_TTS)),
requestExtractor.SetModelAndConfig(func() schema.LocalAIRequest { return new(schema.TTSRequest) }),
localai.TTSEndpoint(cl, ml, appConfig))
vadChain := []fiber.Handler{
requestExtractor.BuildFilteredFirstAvailableDefaultModel(config.BuildUsecaseFilterFn(config.FLAG_VAD)),
requestExtractor.SetModelAndConfig(func() schema.LocalAIRequest { return new(schema.VADRequest) }),
localai.VADEndpoint(cl, ml, appConfig),
}
router.Post("/vad", vadChain...)
router.Post("/v1/vad", vadChain...)
// Stores
sl := model.NewModelLoader("")
@ -47,10 +60,14 @@ func RegisterLocalAIRoutes(router *fiber.App,
router.Get("/metrics", localai.LocalAIMetricsEndpoint())
}
// Experimental Backend Statistics Module
// Backend Statistics Module
// TODO: Should these use standard middlewares? Refactor later, they are extremely simple.
backendMonitorService := services.NewBackendMonitorService(ml, cl, appConfig) // Split out for now
router.Get("/backend/monitor", localai.BackendMonitorEndpoint(backendMonitorService))
router.Post("/backend/shutdown", localai.BackendShutdownEndpoint(backendMonitorService))
// The v1/* urls are exactly the same as above - makes local e2e testing easier if they are registered.
router.Get("/v1/backend/monitor", localai.BackendMonitorEndpoint(backendMonitorService))
router.Post("/v1/backend/shutdown", localai.BackendShutdownEndpoint(backendMonitorService))
// p2p
if p2p.IsP2PEnabled() {
@ -67,6 +84,9 @@ func RegisterLocalAIRoutes(router *fiber.App,
router.Get("/system", localai.SystemInformations(ml, appConfig))
// misc
router.Post("/v1/tokenize", localai.TokenizeEndpoint(cl, ml, appConfig))
router.Post("/v1/tokenize",
requestExtractor.BuildFilteredFirstAvailableDefaultModel(config.BuildUsecaseFilterFn(config.FLAG_TOKENIZE)),
requestExtractor.SetModelAndConfig(func() schema.LocalAIRequest { return new(schema.TokenizeRequest) }),
localai.TokenizeEndpoint(cl, ml, appConfig))
}

125
core/http/routes/openai.go
View file

@ -3,51 +3,50 @@ package routes
import (
"github.com/gofiber/fiber/v2"
"github.com/mudler/LocalAI/core/application"
"github.com/mudler/LocalAI/core/config"
"github.com/mudler/LocalAI/core/http/endpoints/localai"
"github.com/mudler/LocalAI/core/http/endpoints/openai"
"github.com/mudler/LocalAI/core/http/middleware"
"github.com/mudler/LocalAI/core/schema"
)
func RegisterOpenAIRoutes(app *fiber.App,
re *middleware.RequestExtractor,
application *application.Application) {
// openAI compatible API endpoint
// chat
app.Post("/v1/chat/completions",
openai.ChatEndpoint(
application.BackendLoader(),
application.ModelLoader(),
application.TemplatesEvaluator(),
application.ApplicationConfig(),
),
)
app.Post("/chat/completions",
openai.ChatEndpoint(
application.BackendLoader(),
application.ModelLoader(),
application.TemplatesEvaluator(),
application.ApplicationConfig(),
),
)
chatChain := []fiber.Handler{
re.BuildFilteredFirstAvailableDefaultModel(config.BuildUsecaseFilterFn(config.FLAG_CHAT)),
re.SetModelAndConfig(func() schema.LocalAIRequest { return new(schema.OpenAIRequest) }),
re.SetOpenAIRequest,
openai.ChatEndpoint(application.BackendLoader(), application.ModelLoader(), application.TemplatesEvaluator(), application.ApplicationConfig()),
}
app.Post("/v1/chat/completions", chatChain...)
app.Post("/chat/completions", chatChain...)
// edit
app.Post("/v1/edits",
openai.EditEndpoint(
application.BackendLoader(),
application.ModelLoader(),
application.TemplatesEvaluator(),
application.ApplicationConfig(),
),
)
editChain := []fiber.Handler{
re.BuildFilteredFirstAvailableDefaultModel(config.BuildUsecaseFilterFn(config.FLAG_EDIT)),
re.BuildConstantDefaultModelNameMiddleware("gpt-4o"),
re.SetModelAndConfig(func() schema.LocalAIRequest { return new(schema.OpenAIRequest) }),
re.SetOpenAIRequest,
openai.EditEndpoint(application.BackendLoader(), application.ModelLoader(), application.TemplatesEvaluator(), application.ApplicationConfig()),
}
app.Post("/v1/edits", editChain...)
app.Post("/edits", editChain...)
app.Post("/edits",
openai.EditEndpoint(
application.BackendLoader(),
application.ModelLoader(),
application.TemplatesEvaluator(),
application.ApplicationConfig(),
),
)
// completion
completionChain := []fiber.Handler{
re.BuildFilteredFirstAvailableDefaultModel(config.BuildUsecaseFilterFn(config.FLAG_COMPLETION)),
re.BuildConstantDefaultModelNameMiddleware("gpt-4o"),
re.SetModelAndConfig(func() schema.LocalAIRequest { return new(schema.OpenAIRequest) }),
re.SetOpenAIRequest,
openai.CompletionEndpoint(application.BackendLoader(), application.ModelLoader(), application.TemplatesEvaluator(), application.ApplicationConfig()),
}
app.Post("/v1/completions", completionChain...)
app.Post("/completions", completionChain...)
app.Post("/v1/engines/:model/completions", completionChain...)
// assistant
app.Get("/v1/assistants", openai.ListAssistantsEndpoint(application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig()))
@ -81,45 +80,37 @@ func RegisterOpenAIRoutes(app *fiber.App,
app.Get("/v1/files/:file_id/content", openai.GetFilesContentsEndpoint(application.BackendLoader(), application.ApplicationConfig()))
app.Get("/files/:file_id/content", openai.GetFilesContentsEndpoint(application.BackendLoader(), application.ApplicationConfig()))
// completion
app.Post("/v1/completions",
openai.CompletionEndpoint(
application.BackendLoader(),
application.ModelLoader(),
application.TemplatesEvaluator(),
application.ApplicationConfig(),
),
)
app.Post("/completions",
openai.CompletionEndpoint(
application.BackendLoader(),
application.ModelLoader(),
application.TemplatesEvaluator(),
application.ApplicationConfig(),
),
)
app.Post("/v1/engines/:model/completions",
openai.CompletionEndpoint(
application.BackendLoader(),
application.ModelLoader(),
application.TemplatesEvaluator(),
application.ApplicationConfig(),
),
)
// embeddings
app.Post("/v1/embeddings", openai.EmbeddingsEndpoint(application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig()))
app.Post("/embeddings", openai.EmbeddingsEndpoint(application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig()))
app.Post("/v1/engines/:model/embeddings", openai.EmbeddingsEndpoint(application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig()))
embeddingChain := []fiber.Handler{
re.BuildFilteredFirstAvailableDefaultModel(config.BuildUsecaseFilterFn(config.FLAG_EMBEDDINGS)),
re.BuildConstantDefaultModelNameMiddleware("gpt-4o"),
re.SetModelAndConfig(func() schema.LocalAIRequest { return new(schema.OpenAIRequest) }),
re.SetOpenAIRequest,
openai.EmbeddingsEndpoint(application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig()),
}
app.Post("/v1/embeddings", embeddingChain...)
app.Post("/embeddings", embeddingChain...)
app.Post("/v1/engines/:model/embeddings", embeddingChain...)
// audio
app.Post("/v1/audio/transcriptions", openai.TranscriptEndpoint(application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig()))
app.Post("/v1/audio/speech", localai.TTSEndpoint(application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig()))
app.Post("/v1/audio/transcriptions",
re.BuildFilteredFirstAvailableDefaultModel(config.BuildUsecaseFilterFn(config.FLAG_TRANSCRIPT)),
re.SetModelAndConfig(func() schema.LocalAIRequest { return new(schema.OpenAIRequest) }),
re.SetOpenAIRequest,
openai.TranscriptEndpoint(application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig()),
)
app.Post("/v1/audio/speech",
re.BuildFilteredFirstAvailableDefaultModel(config.BuildUsecaseFilterFn(config.FLAG_TTS)),
re.SetModelAndConfig(func() schema.LocalAIRequest { return new(schema.TTSRequest) }),
localai.TTSEndpoint(application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig()))
// images
app.Post("/v1/images/generations", openai.ImageEndpoint(application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig()))
app.Post("/v1/images/generations",
re.BuildConstantDefaultModelNameMiddleware("stablediffusion"),
re.SetModelAndConfig(func() schema.LocalAIRequest { return new(schema.OpenAIRequest) }),
re.SetOpenAIRequest,
openai.ImageEndpoint(application.BackendLoader(), application.ModelLoader(), application.ApplicationConfig()))
if application.ApplicationConfig().ImageDir != "" {
app.Static("/generated-images", application.ApplicationConfig().ImageDir)