feat: add stream events (#152)

.github/workflows/image.yml

@@ -54,8 +54,8 @@ jobs:
         uses: docker/login-action@v2
         with:
           registry: quay.io
-          username: ${{ secrets.QUAY_USERNAME }}
-          password: ${{ secrets.QUAY_PASSWORD }}
+          username: ${{ secrets.LOCALAI_REGISTRY_USERNAME }}
+          password: ${{ secrets.LOCALAI_REGISTRY_PASSWORD }}
       - name: Build
         if: github.event_name != 'pull_request'
         uses: docker/build-push-action@v4

api/openai.go

@@ -2,6 +2,7 @@ package api
 
 import (
 	"bufio"
+	"bytes"
 	"encoding/json"
 	"fmt"
 	"os"
@@ -245,7 +246,7 @@ func completionEndpoint(cm ConfigMerger, debug bool, loader *model.ModelLoader,
 
 		result, err := ComputeChoices(predInput, input, config, loader, func(s string, c *[]Choice) {
 			*c = append(*c, Choice{Text: s})
-		})
+		}, nil)
 		if err != nil {
 			return err
 		}
@@ -290,8 +291,9 @@ func chatEndpoint(cm ConfigMerger, debug bool, loader *model.ModelLoader, thread
 
 		if input.Stream {
 			log.Debug().Msgf("Stream request received")
+			c.Context().SetContentType("text/event-stream")
 			//c.Response().Header.SetContentType(fiber.MIMETextHTMLCharsetUTF8)
-			c.Set("Content-Type", "text/event-stream; charset=utf-8")
+			//	c.Set("Content-Type", "text/event-stream")
 			c.Set("Cache-Control", "no-cache")
 			c.Set("Connection", "keep-alive")
 			c.Set("Transfer-Encoding", "chunked")
@@ -312,53 +314,62 @@ func chatEndpoint(cm ConfigMerger, debug bool, loader *model.ModelLoader, thread
 			log.Debug().Msgf("Template found, input modified to: %s", predInput)
 		}
 
-		result, err := ComputeChoices(predInput, input, config, loader, func(s string, c *[]Choice) {
-			if input.Stream {
-				*c = append(*c, Choice{Delta: &Message{Role: "assistant", Content: s}})
-			} else {
-				*c = append(*c, Choice{Message: &Message{Role: "assistant", Content: s}})
-			}
-		})
-		if err != nil {
-			return err
-		}
-
-		resp := &OpenAIResponse{
-			Model:   input.Model, // we have to return what the user sent here, due to OpenAI spec.
-			Choices: result,
-			Object:  "chat.completion",
-		}
-
 		if input.Stream {
-			resp.Object = "chat.completion.chunk"
-			jsonResult, _ := json.Marshal(resp)
-			log.Debug().Msgf("Response: %s", jsonResult)
-			log.Debug().Msgf("Handling stream request")
+			responses := make(chan OpenAIResponse)
+
+			go func() {
+				ComputeChoices(predInput, input, config, loader, func(s string, c *[]Choice) {}, func(s string) bool {
+					resp := OpenAIResponse{
+						Model:   input.Model, // we have to return what the user sent here, due to OpenAI spec.
+						Choices: []Choice{{Delta: &Message{Role: "assistant", Content: s}}},
+						Object:  "chat.completion.chunk",
+					}
+
+					responses <- resp
+					return true
+				})
+				close(responses)
+			}()
+
 			c.Context().SetBodyStreamWriter(fasthttp.StreamWriter(func(w *bufio.Writer) {
-				fmt.Fprintf(w, "event: data\n")
-				w.Flush()
 
-				fmt.Fprintf(w, "data: %s\n\n", jsonResult)
-				w.Flush()
+				for ev := range responses {
+					var buf bytes.Buffer
+					enc := json.NewEncoder(&buf)
+					enc.Encode(ev)
 
-				fmt.Fprintf(w, "event: data\n")
-				w.Flush()
+					fmt.Fprintf(w, "event: data\n\n")
+					fmt.Fprintf(w, "data: %v\n\n", buf.String())
+					log.Debug().Msgf("Sending chunk: %s", buf.String())
+					w.Flush()
+				}
 
+				w.WriteString("event: data\n\n")
 				resp := &OpenAIResponse{
 					Model:   input.Model, // we have to return what the user sent here, due to OpenAI spec.
 					Choices: []Choice{{FinishReason: "stop"}},
 				}
 				respData, _ := json.Marshal(resp)
 
-				fmt.Fprintf(w, "data: %s\n\n", respData)
+				w.WriteString(fmt.Sprintf("data: %s\n\n", respData))
 				w.Flush()
-
-				//	fmt.Fprintf(w, "data: [DONE]\n\n")
-				//		w.Flush()
 			}))
 			return nil
 		}
 
+		result, err := ComputeChoices(predInput, input, config, loader, func(s string, c *[]Choice) {
+			*c = append(*c, Choice{Message: &Message{Role: "assistant", Content: s}})
+		}, nil)
+		if err != nil {
+			return err
+		}
+
+		resp := &OpenAIResponse{
+			Model:   input.Model, // we have to return what the user sent here, due to OpenAI spec.
+			Choices: result,
+			Object:  "chat.completion",
+		}
+
 		// Return the prediction in the response body
 		return c.JSON(resp)
 	}
@@ -392,7 +403,7 @@ func editEndpoint(cm ConfigMerger, debug bool, loader *model.ModelLoader, thread
 
 		result, err := ComputeChoices(predInput, input, config, loader, func(s string, c *[]Choice) {
 			*c = append(*c, Choice{Text: s})
-		})
+		}, nil)
 		if err != nil {
 			return err
 		}

api/prediction.go

@@ -16,12 +16,12 @@ import (
 var mutexMap sync.Mutex
 var mutexes map[string]*sync.Mutex = make(map[string]*sync.Mutex)
 
-func ModelInference(s string, loader *model.ModelLoader, c Config) (func() (string, error), error) {
+func ModelInference(s string, loader *model.ModelLoader, c Config, tokenCallback func(string) bool) (func() (string, error), error) {
 	var model *llama.LLama
 	var gptModel *gptj.GPTJ
 	var gpt2Model *gpt2.GPT2
 	var stableLMModel *gpt2.StableLM
-
+	supportStreams := false
 	modelFile := c.Model
 
 	// Try to load the model
@@ -125,7 +125,13 @@ func ModelInference(s string, loader *model.ModelLoader, c Config) (func() (stri
 			)
 		}
 	case model != nil:
+		supportStreams = true
 		fn = func() (string, error) {
+
+			if tokenCallback != nil {
+				model.SetTokenCallback(tokenCallback)
+			}
+
 			// Generate the prediction using the language model
 			predictOptions := []llama.PredictOption{
 				llama.SetTemperature(c.Temperature),
@@ -185,11 +191,15 @@ func ModelInference(s string, loader *model.ModelLoader, c Config) (func() (stri
 		l.Lock()
 		defer l.Unlock()
 
-		return fn()
+		res, err := fn()
+		if tokenCallback != nil && !supportStreams {
+			tokenCallback(res)
+		}
+		return res, err
 	}, nil
 }
 
-func ComputeChoices(predInput string, input *OpenAIRequest, config *Config, loader *model.ModelLoader, cb func(string, *[]Choice)) ([]Choice, error) {
+func ComputeChoices(predInput string, input *OpenAIRequest, config *Config, loader *model.ModelLoader, cb func(string, *[]Choice), tokenCallback func(string) bool) ([]Choice, error) {
 	result := []Choice{}
 
 	n := input.N
@@ -199,7 +209,7 @@ func ComputeChoices(predInput string, input *OpenAIRequest, config *Config, load
 	}
 
 	// get the model function to call for the result
-	predFunc, err := ModelInference(predInput, loader, *config)
+	predFunc, err := ModelInference(predInput, loader, *config, tokenCallback)
 	if err != nil {
 		return result, err
 	}