makecindy / makecindy/cindy

自定义模型 deepseek-v4-flash ~50% 概率 400 invalid_parameter_error 阻断任务

Open
#1,981 1 comment 0 reactions 0 assignees View on GitHub
bug
Dominant language
TypeScript
Stars
2.7k
Forks
401
Avg merge
21h 48m
Merged PRs (30d)
776

Description

**提交人**: 匿名
**客户端版本**: 0.1.33

---

## 现象

调用自定义模型 deepseek-v4-flash 时,约 50% 概率出现 `API Error: 400`,错误信息:

```json
{"error":{"code":"invalid_parameter_error","param":null,"message":"'type' must be in [\"enabled\", \"disabled\", \"auto\"]","type":"invalid_request_error"}}
```

该错误会直接阻断当前任务,无法继续。

## 复现步骤

1. 使用自定义模型 deepseek-v4-flash(Cindy 版本 0.1.33)
2. 正常发送消息,agent 执行任务
3. 任务执行到一半时随机出现该错误,约 50% 概率
4. 错误出现后任务被阻断,无法继续

## 期望行为

- 调用自定义模型不应出现 400 参数校验错误
- 该错误不应阻断正在执行的任务,应有重试或降级机制

## 实际行为

API 返回 400 错误,任务被中断,需要手动重试。

## 补充信息

- **Cindy 版本**: 0.1.33
- **模型**: deepseek-v4-flash(自定义模型)
- **触发场景**: 发送消息 / agent 执行任务到一半时出现
- **频率**: 约 50%,会直接阻断任务
- **状态变化**: 昨天之前正常,未修改任何配置,突然出现
---
**版本区域**: CN
**OS**: win32 x64 (10.0.19045)
**界面语言**: zh-CN

Contributor guide

Open the contributing guide

Research direction

Start by tracing the custom-model request path and error handling for the 400 response shown in the report. Reproduce the failure with deepseek-v4-flash on Cindy 0.1.33, then determine what must change so valid requests avoid the invalid type error and the task does not stop without retry or fallback.

Written by the indexing model from the issue text.

Assessment

Tech stack
typescript
Domain
ai, api
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Quiet
Clarity
Needs clarification
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.