[Bug] GEMINI_SUMMARIZE_MODEL

[QAbot-zh]

Bug Description

gemini 系列模型并不都是以gemini-pro为前缀，应该修改GEMINI_SUMMARIZE_MODEL的适用条件，如：currentModel.startsWith("gemini")

function getSummarizeModel(currentModel: string) {
  // if it is using gpt-* models, force to use 3.5 to summarize
  if (currentModel.startsWith("gpt")) {
    return SUMMARIZE_MODEL;
  }
  if (currentModel.startsWith("gemini-pro")) {
    return GEMINI_SUMMARIZE_MODEL;
  }
  return currentModel;
}

Steps to Reproduce

调用 gemini-1.5-pro 模型将不会使用 GEMINI_SUMMARIZE_MODEL 进行总结

Expected Behavior

调用 gemini 系列模型都使用 GEMINI_SUMMARIZE_MODEL 进行总结

Screenshots

No response

Deployment Method

• Docker
• Vercel
• Server

Desktop OS

No response

Desktop Browser

No response

Desktop Browser Version

No response

Smartphone Device

No response

Smartphone OS

No response

Smartphone Browser

No response

Smartphone Browser Version

No response

Additional Logs

No response

[Dean-YZG]

可以说一下原因么？

[Issues-translate-bot]

Bot detected the issue body's language is not English, translate it automatically.

---

Can you tell me why?

[QAbot-zh]

可以说一下原因么？

因为 gemini 的模型名称并不是都以 “gemini-pro” 开头，早期只有gemini-pro和gemini-pro-vision，但现在新模型已经不符合这个规律了

设置 GEMINI_SUMMARIZE_MODEL 本意应该是使用它来对所有的 gemini 模型的会话进行总结对吧，所以应该根据新的模型命名调整判断条件了。实际上 Nextchat 在其他判定是否gemini类型的地方其实已经修改过了，只是总结模型这里留了一个历史遗留问题。

[Issues-translate-bot]

Bot detected the issue body's language is not English, translate it automatically.

---

Can you tell me the reason?

Because not all gemini model names start with "gemini-pro". In the early days, there were only gemini-pro and gemini-pro-vision, but now the new models no longer comply with this rule.

The original intention of setting GEMINI_SUMMARIZE_MODEL is to use it to summarize all gemini model sessions, right, so the judgment conditions should be adjusted according to the new model naming. In fact, Nextchat has actually modified it in other places to determine whether it is a gemini type, but it only leaves a historical issue in the summary model.

[Dean-YZG]

#4625