springAI集成智谱，输入文字图片(多模态)

 API参考 https://www.cnblogs.com/LackyQueen/p/19238256

 

public String multiModalChat(String textQuestion, String imageUrl) {
    List<ChatMessage> messages = new ArrayList<>();
    List<MultiModalContent> contentList = new ArrayList<>();

    // 添加文本内容
    contentList.add(new MultiModalContent(MultiModalContent.Type.TEXT, textQuestion));
    // 添加图片内容
    contentList.add(new MultiModalContent(MultiModalContent.Type.IMAGE_URL, imageUrl));

    // 创建一个包含多模态内容的 UserMessage
    ChatMessage message = new ChatMessage(ChatMessageRole.USER.value(), contentList);
    messages.add(message);

    ChatCompletionRequest request = ChatCompletionRequest.builder()
            .model("glm-4v") // 注意使用多模态模型
            .messages(messages)
            .build();

    ModelApiResponse response = client.invokeModelApi(request);
    return response.getData().getChoices().get(0).getMessage().getContent();
}