adapters/multimodal/client.go

@@ -134,17 +134,25 @@ func (c VisionClient) Describe(ctx context.Context, model, contentType string, d
 	var result struct {
 		Choices []struct {
 			Message struct {
-				Content string `json:"content"`
+				Content          string `json:"content"`
+				ReasoningContent string `json:"reasoning_content"`
 			} `json:"message"`
 		} `json:"choices"`
 	}
 	if err := json.NewDecoder(io.LimitReader(response.Body, 2<<20)).Decode(&result); err != nil {
 		return "", err
 	}
-	if len(result.Choices) == 0 || strings.TrimSpace(result.Choices[0].Message.Content) == "" {
+	if len(result.Choices) == 0 {
 		return "", ErrEmptyResult
 	}
-	return strings.TrimSpace(result.Choices[0].Message.Content), nil
+	text := strings.TrimSpace(result.Choices[0].Message.Content)
+	if text == "" {
+		text = strings.TrimSpace(result.Choices[0].Message.ReasoningContent)
+	}
+	if text == "" {
+		return "", ErrEmptyResult
+	}
+	return text, nil
 }
 
 func (c SpeechClient) client() *http.Client {

adapters/multimodal/client_test.go

@@ -63,3 +63,16 @@ func TestVisionClientSendsImageDataAndInstruction(t *testing.T) {
 		t.Fatalf("description=%q err=%v", text, err)
 	}
 }
+
+func TestVisionClientUsesReasoningContentWhenProviderOmitsContent(t *testing.T) {
+	server := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		_ = json.NewEncoder(w).Encode(map[string]any{"choices": []any{map[string]any{
+			"message": map[string]string{"content": "", "reasoning_content": "чёрный квадрат"},
+		}}})
+	}))
+	defer server.Close()
+	text, err := (VisionClient{BaseURL: server.URL, Client: server.Client()}).Describe(context.Background(), "vision-model", "image/png", []byte("image"), "Опиши изображение")
+	if err != nil || text != "чёрный квадрат" {
+		t.Fatalf("description=%q err=%v", text, err)
+	}
+}