feat: add huggingface embeddings backend

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

Dockerfile

@@ -11,10 +11,15 @@ ARG TARGETARCH
 ARG TARGETVARIANT
 
 ENV BUILD_TYPE=${BUILD_TYPE}
+ENV EXTERNAL_GRPC_BACKENDS="huggingface-embeddings:/build/extra/grpc/huggingface/huggingface.py"
 ARG GO_TAGS="stablediffusion tts"
 
 RUN apt-get update && \
-    apt-get install -y ca-certificates cmake curl patch
+    apt-get install -y ca-certificates cmake curl patch pip
+
+# Extras requirements
+COPY extra/requirements.txt /build/extra/requirements.txt
+RUN pip install -r /build/extra/requirements.txt && rm -rf /build/extra/requirements.txt
 
 # CuBLAS requirements
 RUN if [ "${BUILD_TYPE}" = "cublas" ]; then \

Makefile

@@ -313,7 +313,7 @@ test: prepare test-models/testmodel grpcs
 	@echo 'Running tests'
 	export GO_TAGS="tts stablediffusion"
 	$(MAKE) prepare-test
-	TEST_DIR=$(abspath ./)/test-dir/ FIXTURES=$(abspath ./)/tests/fixtures CONFIG_FILE=$(abspath ./)/test-models/config.yaml MODELS_PATH=$(abspath ./)/test-models \
+	HUGGINGFACE_GRPC=$(abspath ./)/extra/grpc/huggingface/huggingface.py TEST_DIR=$(abspath ./)/test-dir/ FIXTURES=$(abspath ./)/tests/fixtures CONFIG_FILE=$(abspath ./)/test-models/config.yaml MODELS_PATH=$(abspath ./)/test-models \
 	$(GOCMD) run github.com/onsi/ginkgo/v2/ginkgo --label-filter="!gpt4all && !llama" --flake-attempts 5 -v -r ./api ./pkg
 	$(MAKE) test-gpt4all
 	$(MAKE) test-llama
@@ -338,9 +338,7 @@ test-stablediffusion: prepare-test
 
 test-container:
 	docker build --target requirements -t local-ai-test-container .
-	docker run --name localai-tests -e GO_TAGS=$(GO_TAGS) -ti -v $(abspath ./):/build local-ai-test-container make test
-	docker rm localai-tests
-	docker rmi local-ai-test-container
+	docker run -ti --rm --entrypoint /bin/bash -ti -v $(abspath ./):/build local-ai-test-container
 
 ## Help:
 help: ## Show this help.
@@ -354,10 +352,15 @@ help: ## Show this help.
 		else if (/^## .*$$/) {printf "  ${CYAN}%s${RESET}\n", substr($$1,4)} \
 		}' $(MAKEFILE_LIST)
 
-protogen:
+protogen: protogen-go protogen-python
+
+protogen-go:
 	protoc --go_out=. --go_opt=paths=source_relative --go-grpc_out=. --go-grpc_opt=paths=source_relative \
     pkg/grpc/proto/backend.proto
 
+protogen-python:
+	python -m grpc_tools.protoc -Ipkg/grpc/proto/ --python_out=extra/grpc/huggingface/ --grpc_python_out=extra/grpc/huggingface/ pkg/grpc/proto/backend.proto
+
 ## GRPC
 
 backend-assets/grpc:

api/api_test.go

@@ -386,6 +386,102 @@ var _ = Describe("API test", func() {
 		})
 	})
 
+	Context("External gRPCs", func() {
+		BeforeEach(func() {
+			modelLoader = model.NewModelLoader(os.Getenv("MODELS_PATH"))
+			c, cancel = context.WithCancel(context.Background())
+
+			app, err := App(
+				append(commonOpts,
+					options.WithContext(c),
+					options.WithAudioDir(tmpdir),
+					options.WithImageDir(tmpdir),
+					options.WithModelLoader(modelLoader),
+					options.WithBackendAssets(backendAssets),
+					options.WithExternalBackend("huggingface", os.Getenv("HUGGINGFACE_GRPC")),
+					options.WithBackendAssetsOutput(tmpdir))...,
+			)
+			Expect(err).ToNot(HaveOccurred())
+			go app.Listen("127.0.0.1:9090")
+
+			defaultConfig := openai.DefaultConfig("")
+			defaultConfig.BaseURL = "http://127.0.0.1:9090/v1"
+
+			// Wait for API to be ready
+			client = openai.NewClientWithConfig(defaultConfig)
+			Eventually(func() error {
+				_, err := client.ListModels(context.TODO())
+				return err
+			}, "2m").ShouldNot(HaveOccurred())
+		})
+
+		AfterEach(func() {
+			cancel()
+			app.Shutdown()
+			os.RemoveAll(tmpdir)
+		})
+
+		Context("API query", func() {
+			BeforeEach(func() {
+				modelLoader = model.NewModelLoader(os.Getenv("MODELS_PATH"))
+				c, cancel = context.WithCancel(context.Background())
+
+				var err error
+				app, err = App(
+					append(commonOpts,
+						options.WithDebug(true),
+						options.WithContext(c), options.WithModelLoader(modelLoader))...)
+				Expect(err).ToNot(HaveOccurred())
+				go app.Listen("127.0.0.1:9090")
+
+				defaultConfig := openai.DefaultConfig("")
+				defaultConfig.BaseURL = "http://127.0.0.1:9090/v1"
+
+				client2 = openaigo.NewClient("")
+				client2.BaseURL = defaultConfig.BaseURL
+
+				// Wait for API to be ready
+				client = openai.NewClientWithConfig(defaultConfig)
+				Eventually(func() error {
+					_, err := client.ListModels(context.TODO())
+					return err
+				}, "2m").ShouldNot(HaveOccurred())
+			})
+			AfterEach(func() {
+				cancel()
+				app.Shutdown()
+			})
+
+			It("calculate embeddings with huggingface", func() {
+				if runtime.GOOS != "linux" {
+					Skip("test supported only on linux")
+				}
+				resp, err := client.CreateEmbeddings(
+					context.Background(),
+					openai.EmbeddingRequest{
+						Model: openai.AdaCodeSearchCode,
+						Input: []string{"sun", "cat"},
+					},
+				)
+				Expect(err).ToNot(HaveOccurred())
+				Expect(len(resp.Data[0].Embedding)).To(BeNumerically("==", 384))
+				Expect(len(resp.Data[1].Embedding)).To(BeNumerically("==", 384))
+
+				sunEmbedding := resp.Data[0].Embedding
+				resp2, err := client.CreateEmbeddings(
+					context.Background(),
+					openai.EmbeddingRequest{
+						Model: openai.AdaCodeSearchCode,
+						Input: []string{"sun"},
+					},
+				)
+				Expect(err).ToNot(HaveOccurred())
+				Expect(resp2.Data[0].Embedding).To(Equal(sunEmbedding))
+				Expect(resp2.Data[0].Embedding).ToNot(Equal(resp.Data[1].Embedding))
+			})
+		})
+	})
+
 	Context("Model gallery", func() {
 		BeforeEach(func() {
 			var err error
@@ -530,7 +626,7 @@ var _ = Describe("API test", func() {
 		It("returns the models list", func() {
 			models, err := client.ListModels(context.TODO())
 			Expect(err).ToNot(HaveOccurred())
-			Expect(len(models.Models)).To(Equal(10))
+			Expect(len(models.Models)).To(Equal(11))
 		})
 		It("can generate completions", func() {
 			resp, err := client.CreateCompletion(context.TODO(), openai.CompletionRequest{Model: "testmodel", Prompt: "abcdedfghikl"})
@@ -707,7 +803,7 @@ var _ = Describe("API test", func() {
 		It("can generate chat completions from config file", func() {
 			models, err := client.ListModels(context.TODO())
 			Expect(err).ToNot(HaveOccurred())
-			Expect(len(models.Models)).To(Equal(12))
+			Expect(len(models.Models)).To(Equal(13))
 		})
 		It("can generate chat completions from config file", func() {
 			resp, err := client.CreateChatCompletion(context.TODO(), openai.ChatCompletionRequest{Model: "list1", Messages: []openai.ChatCompletionMessage{openai.ChatCompletionMessage{Role: "user", Content: "abcdedfghikl"}}})