diff --git a/relay/adaptor/openai/adaptor.go b/relay/adaptor/openai/adaptor.go
index 5dc395ad..6946e402 100644
--- a/relay/adaptor/openai/adaptor.go
+++ b/relay/adaptor/openai/adaptor.go
@@ -75,6 +75,13 @@ func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *model.G
 	if request == nil {
 		return nil, errors.New("request is nil")
 	}
+	if request.Stream {
+		// always return usage in stream mode
+		if request.StreamOptions == nil {
+			request.StreamOptions = &model.StreamOptions{}
+		}
+		request.StreamOptions.IncludeUsage = true
+	}
 	return request, nil
 }
 
diff --git a/relay/model/constant.go b/relay/model/constant.go
index f6cf1924..c9d6d645 100644
--- a/relay/model/constant.go
+++ b/relay/model/constant.go
@@ -1,6 +1,7 @@
 package model
 
 const (
-	ContentTypeText     = "text"
-	ContentTypeImageURL = "image_url"
+	ContentTypeText       = "text"
+	ContentTypeImageURL   = "image_url"
+	ContentTypeInputAudio = "input_audio"
 )
diff --git a/relay/model/general.go b/relay/model/general.go
index aacc8467..fe73779e 100644
--- a/relay/model/general.go
+++ b/relay/model/general.go
@@ -12,9 +12,20 @@ type JSONSchema struct {
 	Strict      *bool                  `json:"strict,omitempty"`
 }
 
+type Audio struct {
+	Voice  string `json:"voice,omitempty"`
+	Format string `json:"format,omitempty"`
+}
+
+type StreamOptions struct {
+	IncludeUsage bool `json:"include_usage,omitempty"`
+}
+
 type GeneralOpenAIRequest struct {
 	Messages         []Message       `json:"messages,omitempty"`
 	Model            string          `json:"model,omitempty"`
+	Modalities       []string        `json:"modalities,omitempty"`
+	Audio            *Audio          `json:"audio,omitempty"`
 	FrequencyPenalty float64         `json:"frequency_penalty,omitempty"`
 	MaxTokens        int             `json:"max_tokens,omitempty"`
 	N                int             `json:"n,omitempty"`
@@ -23,6 +34,7 @@ type GeneralOpenAIRequest struct {
 	Seed             float64         `json:"seed,omitempty"`
 	Stop             any             `json:"stop,omitempty"`
 	Stream           bool            `json:"stream,omitempty"`
+	StreamOptions    *StreamOptions  `json:"stream_options,omitempty"`
 	Temperature      float64         `json:"temperature,omitempty"`
 	TopP             float64         `json:"top_p,omitempty"`
 	TopK             int             `json:"top_k,omitempty"`
@@ -37,7 +49,7 @@ type GeneralOpenAIRequest struct {
 	Dimensions       int             `json:"dimensions,omitempty"`
 	Instruction      string          `json:"instruction,omitempty"`
 	Size             string          `json:"size,omitempty"`
-	NumCtx           int         	 `json:"num_ctx,omitempty"`
+	NumCtx           int             `json:"num_ctx,omitempty"`
 }
 
 func (r GeneralOpenAIRequest) ParseInput() []string {