refinements

mudler · Feb 7, 2024 · efcf193 · efcf193
1 parent a7a1038
commit efcf193
Show file tree

Hide file tree

Showing 3 changed files with 18 additions and 14 deletions.
diff --git a/Dockerfile b/Dockerfile
@@ -5,7 +5,6 @@ ARG IMAGE_TYPE=extras
 FROM ubuntu:22.04 as requirements-core
 
 ARG GO_VERSION=1.21.7
-ARG GO_ARCH=amd64
 ARG BUILD_TYPE
 ARG CUDA_MAJOR_VERSION=11
 ARG CUDA_MINOR_VERSION=7
@@ -21,10 +20,8 @@ ARG GO_TAGS="stablediffusion tinydream tts"
 RUN apt-get update && \
     apt-get install -y ca-certificates curl patch pip cmake git && apt-get clean
 
-# Download Go 1.2.2 and install it to /usr/local/go
-RUN curl -L -s https://go.dev/dl/go$GO_VERSION.linux-$GO_ARCH.tar.gz | tar -v -C /usr/local -xz
-
-# Let's people find our Go binaries
+# Install Go
+RUN curl -L -s https://go.dev/dl/go$GO_VERSION.linux-$TARGETARCH.tar.gz | tar -v -C /usr/local -xz
 ENV PATH $PATH:/usr/local/go/bin
 
 COPY --chmod=644 custom-ca-certs/* /usr/local/share/ca-certificates/

diff --git a/Makefile b/Makefile
@@ -148,7 +148,8 @@ ALL_GRPC_BACKENDS+=backend-assets/grpc/llama
 ALL_GRPC_BACKENDS+=backend-assets/grpc/llama-cpp
 ALL_GRPC_BACKENDS+=backend-assets/grpc/llama-ggml
 ALL_GRPC_BACKENDS+=backend-assets/grpc/gpt4all
-ALL_GRPC_BACKENDS+=backend-assets/grpc/rwkv backend-assets/grpc/whisper
+ALL_GRPC_BACKENDS+=backend-assets/grpc/rwkv
+ALL_GRPC_BACKENDS+=backend-assets/grpc/whisper
 ALL_GRPC_BACKENDS+=$(OPTIONAL_GRPC)
 
 GRPC_BACKENDS?=$(ALL_GRPC_BACKENDS) $(OPTIONAL_GRPC)

diff --git a/api/api_test.go b/api/api_test.go
@@ -29,6 +29,12 @@ import (
 	"github.com/sashabaranov/go-openai/jsonschema"
 )
 
+const testPrompt = `Below is an instruction that describes a task. Write a response that appropriately completes the request.
+
+### Instruction: Describe how to make a cup of coffee
+
+### Response:`
+
 type modelApplyRequest struct {
 	ID        string                 `json:"id"`
 	URL       string                 `json:"url"`
@@ -460,7 +466,7 @@ var _ = Describe("API test", func() {
 					return response["processed"].(bool)
 				}, "960s", "10s").Should(Equal(true))
 
-				resp, err := client.CreateChatCompletion(context.TODO(), openai.ChatCompletionRequest{Model: "gpt4all-j", Messages: []openai.ChatCompletionMessage{openai.ChatCompletionMessage{Role: "user", Content: "How are you?"}}})
+				resp, err := client.CreateChatCompletion(context.TODO(), openai.ChatCompletionRequest{Model: "gpt4all-j", Messages: []openai.ChatCompletionMessage{openai.ChatCompletionMessage{Role: "user", Content: testPrompt}}})
 				Expect(err).ToNot(HaveOccurred())
 				Expect(len(resp.Choices)).To(Equal(1))
 				Expect(resp.Choices[0].Message.Content).To(ContainSubstring("well"))
@@ -629,36 +635,36 @@ var _ = Describe("API test", func() {
 			Expect(len(models.Models)).To(Equal(6)) // If "config.yaml" should be included, this should be 8?
 		})
 		It("can generate completions", func() {
-			resp, err := client.CreateCompletion(context.TODO(), openai.CompletionRequest{Model: "testmodel", Prompt: "abcdedfghikl"})
+			resp, err := client.CreateCompletion(context.TODO(), openai.CompletionRequest{Model: "testmodel", Prompt: testPrompt})
 			Expect(err).ToNot(HaveOccurred())
 			Expect(len(resp.Choices)).To(Equal(1))
 			Expect(resp.Choices[0].Text).ToNot(BeEmpty())
 		})
 
 		It("can generate chat completions ", func() {
-			resp, err := client.CreateChatCompletion(context.TODO(), openai.ChatCompletionRequest{Model: "testmodel", Messages: []openai.ChatCompletionMessage{openai.ChatCompletionMessage{Role: "user", Content: "abcdedfghikl"}}})
+			resp, err := client.CreateChatCompletion(context.TODO(), openai.ChatCompletionRequest{Model: "testmodel", Messages: []openai.ChatCompletionMessage{openai.ChatCompletionMessage{Role: "user", Content: testPrompt}}})
 			Expect(err).ToNot(HaveOccurred())
 			Expect(len(resp.Choices)).To(Equal(1))
 			Expect(resp.Choices[0].Message.Content).ToNot(BeEmpty())
 		})
 
 		It("can generate completions from model configs", func() {
-			resp, err := client.CreateCompletion(context.TODO(), openai.CompletionRequest{Model: "gpt4all", Prompt: "abcdedfghikl"})
+			resp, err := client.CreateCompletion(context.TODO(), openai.CompletionRequest{Model: "gpt4all", Prompt: testPrompt})
 			Expect(err).ToNot(HaveOccurred())
 			Expect(len(resp.Choices)).To(Equal(1))
 			Expect(resp.Choices[0].Text).ToNot(BeEmpty())
 		})
 
 		It("can generate chat completions from model configs", func() {
-			resp, err := client.CreateChatCompletion(context.TODO(), openai.ChatCompletionRequest{Model: "gpt4all-2", Messages: []openai.ChatCompletionMessage{openai.ChatCompletionMessage{Role: "user", Content: "abcdedfghikl"}}})
+			resp, err := client.CreateChatCompletion(context.TODO(), openai.ChatCompletionRequest{Model: "gpt4all-2", Messages: []openai.ChatCompletionMessage{openai.ChatCompletionMessage{Role: "user", Content: testPrompt}}})
 			Expect(err).ToNot(HaveOccurred())
 			Expect(len(resp.Choices)).To(Equal(1))
 			Expect(resp.Choices[0].Message.Content).ToNot(BeEmpty())
 		})
 
 		It("returns errors", func() {
 			backends := len(model.AutoLoadBackends) + 1 // +1 for huggingface
-			_, err := client.CreateCompletion(context.TODO(), openai.CompletionRequest{Model: "foomodel", Prompt: "abcdedfghikl"})
+			_, err := client.CreateCompletion(context.TODO(), openai.CompletionRequest{Model: "foomodel", Prompt: testPrompt})
 			Expect(err).To(HaveOccurred())
 			Expect(err.Error()).To(ContainSubstring(fmt.Sprintf("error, status code: 500, message: could not load model - all backends returned error: %d errors occurred:", backends)))
 		})
@@ -834,13 +840,13 @@ var _ = Describe("API test", func() {
 			app.Shutdown()
 		})
 		It("can generate chat completions from config file (list1)", func() {
-			resp, err := client.CreateChatCompletion(context.TODO(), openai.ChatCompletionRequest{Model: "list1", Messages: []openai.ChatCompletionMessage{{Role: "user", Content: "abcdedfghikl"}}})
+			resp, err := client.CreateChatCompletion(context.TODO(), openai.ChatCompletionRequest{Model: "list1", Messages: []openai.ChatCompletionMessage{{Role: "user", Content: testPrompt}}})
 			Expect(err).ToNot(HaveOccurred())
 			Expect(len(resp.Choices)).To(Equal(1))
 			Expect(resp.Choices[0].Message.Content).ToNot(BeEmpty())
 		})
 		It("can generate chat completions from config file (list2)", func() {
-			resp, err := client.CreateChatCompletion(context.TODO(), openai.ChatCompletionRequest{Model: "list2", Messages: []openai.ChatCompletionMessage{{Role: "user", Content: "abcdedfghikl"}}})
+			resp, err := client.CreateChatCompletion(context.TODO(), openai.ChatCompletionRequest{Model: "list2", Messages: []openai.ChatCompletionMessage{{Role: "user", Content: testPrompt}}})
 			Expect(err).ToNot(HaveOccurred())
 			Expect(len(resp.Choices)).To(Equal(1))
 			Expect(resp.Choices[0].Message.Content).ToNot(BeEmpty())