结合Gemini 1.5 flash使用，模型并不会总结工具或知识库获取到的信息

[baiyi-os]

Self Checks

• [X] This is only for bug report, if you would like to ask a question, please head to Discussions.
• [X] I have searched for existing issues search for existing issues, including closed ones.
• [X] I confirm that I am using English to submit this report (我已阅读并同意 Language Policy).
• [X] [FOR CHINESE USERS] 请务必使用英文提交 Issue，否则会被关闭。谢谢！:）
• [X] Please do not modify this template :) and fill in all the required fields.

Dify version

0.8.2

Cloud or Self Hosted

Self Hosted (Docker)

Steps to reproduce

选择Gemini 1.5 flash，添加知识库或工具提问即可

✔️ Expected Behavior

No response

❌ Actual Behavior

[dosubot[bot]]

@baiyi-os 你好! 我是Dosu，一个可以帮助解决错误、回答问题并成为贡献者的机器人。我会在等待人类维护者的过程中为你提供帮助。

I found a similar unsolved discussion that might be relevant to your issue: Why this not work, which was last updated on September 5, 2024 ^[1].

Regarding your issue, it appears that the Gemini 1.5 flash model is not summarizing information retrieved from tools or knowledge bases as expected. Unfortunately, I couldn't find specific documentation or example code that demonstrates how to effectively use the Gemini 1.5 flash model for summarizing information from knowledge bases within the repository ^{[2][3][4][5][6]}.

You might want to check the configuration steps for setting up the Gemini 1.5 flash model in a self-hosted environment to ensure everything is correctly configured. Here are the steps:

model: gemini-1.5-flash-001
label:
  en_US: Gemini 1.5 Flash 001
model_type: llm
features:
  - agent-thought
  - vision
  - tool-call
  - stream-tool-call
model_properties:
  mode: chat
  context_size: 1048576
parameter_rules:
  - name: temperature
    use_template: temperature
  - name: top_p
    use_template: top_p
  - name: top_k
    label:
      zh_Hans: 取样数量
      en_US: Top k
    type: int
    help:
      zh_Hans: 仅从每个后续标记的前 K 个选项中采样。
      en_US: Only sample from the top K options for each subsequent token.
    required: false
  - name: max_tokens_to_sample
    use_template: max_tokens
    required: true
    default: 8192
    min: 1
    max: 8192
  - name: response_format
    use_template: response_format
  - name: stream
    label:
      zh_Hans: 流式输出
      en_US: Stream
    type: boolean
    help:
      zh_Hans: 流式输出允许模型在生成文本的过程中逐步返回结果，而不是一次性生成全部结果后再返回。
      en_US: Streaming output allows the model to return results incrementally as it generates text, rather than generating all the results at once.
    default: false
pricing:
  input: '0.00'
  output: '0.00'
  unit: '0.000001'
  currency: USD

Make sure to adjust the parameters according to your specific requirements ^[2][6][7][4].

If the issue persists, you might want to explore further discussions or issues in the repository for more insights.

To continue talking to Dosu, mention @dosu.