和oneapi结合一直显示模型无可用渠道

[flyinsz]

如题，我已再三确认模型是存在的。

配置如下：

openai:

OpenAI 相关配置。OpenAI管理后台: https://platform.openai.com/api-keys

#api_key:
#base_url: # API的网址。原版API使用默认值: https://api.openai.com/v1
api_key: sk-zytscKkrg06aExzG41F898Ef82F046Ea8bAd51Bb82D5280f
base_url: http://localhost:3002/v1
proxy: # 代理服务器地址。格式为"http://地址:端口号"
chat_model: gpt-3.5-turbo # 聊天使用模型名称，例如：gpt-3.5-turbo-0125, gpt-4-turbo-preview. 可用模型参见(https://platform.openai.com/docs/models/overview)

作图参数, 参见(https://cookbook.openai.com/articles/what_is_new_with_dalle_3) 以及 api 说明

image_model: dall-e-3 # (可选, 默认值: dall-e-3) 作图模型, 选项: dall-e-3, dall-e-2
image_quality: hd # 现在作图质量由模型决定, 该参数暂时无效。(可选, 默认值: standard) 作图质量。选项: standard(标准), hd(高清)。
image_size: 1024x1024 # (可选, 默认值: 1024x1024) 作图大小, dall-e-3 支持尺寸 1024x1024, 1792x1024, 1024x1792。

语音参数. 参考: https://platform.openai.com/docs/api-reference/audio/createSpeech

voice: alloy # 虚拟配音演员 alloy, echo, fable, onyx, nova, and shimmer. 试听: https://platform.openai.com/docs/guides/text-to-speech/voice-options
voice_speed: 1.1 # 语音速度, 可选 0.25-4.0 (默认1.0)
transcript_prompt: "请转录用户的语音消息，可能使用的语言是：简体中文和English" # 为语音转录提供一段提示词，以提高识别正确率

错误如下：
2024-05-05 23:52:18 初始化OpenAI API...
2024-05-05 23:52:18 启用工具 image_to_text (获取图像的文字描述)
2024-05-05 23:52:18 启用工具 text_to_image (用文字描述生成图像)
2024-05-05 23:52:18 启用工具 text_to_speech (用文字生成语音)
2024-05-05 23:52:18 启用工具 browse_link (访问链接获取内容)
2024-05-05 23:52:18 启用工具 bing_search (用必应搜索网络内容)
2024-05-05 23:52:18 启用工具 audio_transcript (语音转录成文字)
2024-05-05 23:52:18 启用工具 video_analysis (分析视频内容)
主程序发生错误: Traceback (most recent call last):
File "main.py", line 42, in
File "main.py", line 16, in main
File "openai_wrapper.py", line 42, in init
File "openai_wrapper.py", line 65, in load_config
File "openai_wrapper.py", line 132, in get_assistant
File "openai\resources\beta\assistants\assistants.py", line 287, in list
File "openai_base_client.py", line 1249, in get_api_list
File "openai_base_client.py", line 1094, in _request_api_list
File "openai_base_client.py", line 889, in request
File "openai_base_client.py", line 965, in _request
File "openai_base_client.py", line 1013, in _retry_request
File "openai_base_client.py", line 965, in _request
File "openai_base_client.py", line 1013, in _retry_request
File "openai_base_client.py", line 980, in _request
openai.InternalServerError: Error code: 503 - {'error': {'message': '当前分组 default 下对于模型 无可用渠道 (request id: 2024050523522150057096679513566)', 'type': 'one_api_error'}}

2024-05-05 23:52:21 主程序发生错误, 即将退出: Traceback (most recent call last):
File "main.py", line 42, in
File "main.py", line 16, in main
File "openai_wrapper.py", line 42, in init
File "openai_wrapper.py", line 65, in load_config
File "openai_wrapper.py", line 132, in get_assistant
File "openai\resources\beta\assistants\assistants.py", line 287, in list
File "openai_base_client.py", line 1249, in get_api_list
File "openai_base_client.py", line 1094, in _request_api_list
File "openai_base_client.py", line 889, in request
File "openai_base_client.py", line 965, in _request
File "openai_base_client.py", line 1013, in _retry_request
File "openai_base_client.py", line 965, in _request
File "openai_base_client.py", line 1013, in _retry_request
File "openai_base_client.py", line 980, in _request
openai.InternalServerError: Error code: 503 - {'error': {'message': '当前分组 default 下对于模型 无可用渠道 (request id: 2024050523522150057096679513566)', 'type': 'one_api_error'}}

[leeileduo]

labring/FastGPT#2105