Skip to content

gpt-5.3-codex模型疑似被投毒,输出菠菜内容,有佬有解决办法嘛 #629

@LUXINGYU23

Description

@LUXINGYU23
Image Image Image 现在这个问题在多个项目内都频繁出现了,导致基础工具调用基本上都失败,已经基本无法使用该模型了,目前只能退而求其次使用gpt5.2,没有这个问题

接入到opencode里使用,模型配置如下
"gpt-5.3-codex": {
"name": "GPT-5.3 Codex",
"thinking": true,
"modalities": {
"input": [
"text",
"image",
"pdf"
],
"output": [
"text"
]
},
"limit": {
"context": 200000,
"output": 128000
},
"options": {
"store": false,
"reasoningEffort": "xhigh",
"textVerbosity": "high",
"reasoningSummary": "auto",
"include": [
"reasoning.encrypted_content"
]
}
},

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions