groq经常返回{'error': "error, status code: 403, message: invalid character '<' looking for beginning of value"}

[botoai]

同样的配置， F:\simple-one-api>python openai.py {'error': "error, status code: 403, message: invalid character '<' looking for beginning of value"} Traceback (most recent call last): File "F:\simple-one-api\openai.py", line 5, in import openai File "F:\simple-one-api\openai.py", line 48, in message = result['error']['message'] TypeError: string indices must be integers

F:\simple-one-api>python openai.py {'id': 'as-f58fyz6321', 'object': 'chat.completion', 'created': 1719405301, 'model': 'ERNIE-Tiny-8K', 'choices': [{'index': 0, 'message': {'role': 'assistant', 'content': '我是由人工智能技术开发的模型，属于一个开源的机器学习平台。我来自多个公司，但并没有具体的公司背景。如果您需要关于特定模型或公司的信息，建议您访问相关的官方网站或社交媒体平台。'}, 'logprobs': None, 'finish_reason': 'completed'}], 'usage': {'prompt_tokens': 7, 'completion_tokens': 45, 'total_tokens': 52}}

配置如下 { "models": [ "llama3-70b-8192", "llama3-8b-8192", "gemma-7b-it", "mixtral-8x7b-32768" ], "enabled": true, "credentials": { "api_key": "gsk_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx" }, "server_url": "https://api.groq.com/openai/v1" }

[botoai]

有一会配置又是正常的，但不记得怎么弄出来的了，其他的baidu,tencent,xinghuo大模型都正常，就这个一直这样

[WBinBin001]

有一个配置又是正常的，但不记得怎么弄出来的了，其他的baidu,tencent,xinghuo大模型都正常，就这个这个样

你改好了么，我也一样，怎么都配不好。。。。

[fruitbars]

同样的配置， F:\simple-one-api>python openai.py {'error': "error, status code: 403, message: invalid character '<' looking for beginning of value"} Traceback (most recent call last): File "F:\simple-one-api\openai.py", line 5, in import openai File "F:\simple-one-api\openai.py", line 48, in message = result['error']['message'] TypeError: string indices must be integers

F:\simple-one-api>python openai.py {'id': 'as-f58fyz6321', 'object': 'chat.completion', 'created': 1719405301, 'model': 'ERNIE-Tiny-8K', 'choices': [{'index': 0, 'message': {'role': 'assistant', 'content': '我是由人工智能技术开发的模型，属于一个开源的机器学习平台。我来自多个公司，但并没有具体的公司背景。如果您需要关于特定模型或公司的信息，建议您访问相关的官方网站或社交媒体平台。'}, 'logprobs': None, 'finish_reason': 'completed'}], 'usage': {'prompt_tokens': 7, 'completion_tokens': 45, 'total_tokens': 52}}

配置如下 { "models": [ "llama3-70b-8192", "llama3-8b-8192", "gemma-7b-it", "mixtral-8x7b-32768" ], "enabled": true, "credentials": { "api_key": "gsk_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx" }, "server_url": "https://api.groq.com/openai/v1" }

openai.py 这个脚本可以将返回的http body打印一下出来 错误信息吗？

[botoai]

同样的配置， F:\simple-one-api>python openai.py {'error': "error, status code: 403, message: invalid character '<' looking for beginning of value"} Traceback (most recent call last): File "F:\simple-one-api\openai.py", line 5, in import openai File "F:\simple-one-api\openai.py", line 48, in message = result['error']['message'] TypeError: string indices must be integers F:\simple-one-api>python openai.py {'id': 'as-f58fyz6321', 'object': 'chat.completion', 'created': 1719405301, 'model': 'ERNIE-Tiny-8K', 'choices': [{'index': 0, 'message': {'role': 'assistant', 'content': '我是由人工智能技术开发的模型，属于一个开源的机器学习平台。我来自多个公司，但并没有具体的公司背景。如果您需要关于特定模型或公司的信息，建议您访问相关的官方网站或社交媒体平台。'}, 'logprobs': None, 'finish_reason': 'completed'}], 'usage': {'prompt_tokens': 7, 'completion_tokens': 45, 'total_tokens': 52}} 配置如下 { "models": [ "llama3-70b-8192", "llama3-8b-8192", "gemma-7b-it", "mixtral-8x7b-32768" ], "enabled": true, "credentials": { "api_key": "gsk_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx" }, "server_url": "https://api.groq.com/openai/v1" }

openai.py 这个脚本可以将返回的http body打印一下出来 错误信息吗？

有一个配置又是正常的，但不记得怎么弄出来的了，其他的baidu,tencent,xinghuo大模型都正常，就这个这个样

你改好了么，我也一样，怎么都配不好。。。。

没有

[botoai]

同样的配置， F:\simple-one-api>python openai.py {'error': "error, status code: 403, message: invalid character '<' looking for beginning of value"} Traceback (most recent call last): File "F:\simple-one-api\openai.py", line 5, in import openai File "F:\simple-one-api\openai.py", line 48, in message = result['error']['message'] TypeError: string indices must be integers F:\simple-one-api>python openai.py {'id': 'as-f58fyz6321', 'object': 'chat.completion', 'created': 1719405301, 'model': 'ERNIE-Tiny-8K', 'choices': [{'index': 0, 'message': {'role': 'assistant', 'content': '我是由人工智能技术开发的模型，属于一个开源的机器学习平台。我来自多个公司，但并没有具体的公司背景。如果您需要关于特定模型或公司的信息，建议您访问相关的官方网站或社交媒体平台。'}, 'logprobs': None, 'finish_reason': 'completed'}], 'usage': {'prompt_tokens': 7, 'completion_tokens': 45, 'total_tokens': 52}} 配置如下 { "models": [ "llama3-70b-8192", "llama3-8b-8192", "gemma-7b-it", "mixtral-8x7b-32768" ], "enabled": true, "credentials": { "api_key": "gsk_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx" }, "server_url": "https://api.groq.com/openai/v1" }

openai.py 这个脚本可以将返回的http body打印一下出来 错误信息吗？

{"error":"error, status code: 403, message: invalid character '\u003c' looking for beginning of value"} {'error': "error, status code: 403, message: invalid character '<' looking for beginning of value"} Traceback (most recent call last): File "F:\simple-one-api\simple-oneapi.py", line 46, in print("\r\n", result["choices"][0]["message"]["content"].strip()) KeyError: 'choices'

[fruitbars]

这是json解析时候报错，解析之前收到的http body的数据能print出来吗？

[botoai]

这是json解析时候报错，解析之前收到的http body的数据能print出来吗？ 怎么打印法？或者要怎么配置？

我代码 `apikey = "sk-123456" #4

设置 API 密钥和 API URL

url = "http://192.168.1.32:3035/v1/chat/completions"

设置请求头

headers = { "Content-Type": "application/json", "Authorization": "Bearer " + apikey, }

设置请求体

prompt = "which model are you? who developed you?" field = { "model":"llama3-70b-8192",

"model":"Atom-13B-Chat",

"messages":[
    {"role": "user", "content": prompt},
]

}

发送 HTTP POST 请求

response = requests.post(url, headers=headers, data=json.dumps(field))

解析响应结果

result = response.json() print(result) print("\r\n", result["choices"][0]["message"]["content"].strip())`

[fruitbars]

类似这样吧，👀看具体输出的消息内容

response = requests.post(url, headers=headers, data=json.dumps(field))

if response.status_code != 200: print(response.text) else:

解析response

pass

[botoai]

response = requests.post(url, headers=headers, data=json.dumps(field)) print(response.text) print(response.json())

返回 {"error":"error, status code: 403, message: invalid character '\u003c' looking for beginning of value"} {'error': "error, status code: 403, message: invalid character '<' looking for beginning of value"}

[botoai]

2024-06-27T12:19:01.133+0800 ERROR handler/openai_openai_handler.go:157 An error occurred {"req": {"model":"llama3-70b-8192","messages":[{"role":"user","content":"1+1=?"}],"temperature":0.1}, "error": "error, status code: 403, message: invalid character '<' looking for beginning of value"}

[Blackwatch927]

json解析报错，回传的HTML可以看一下，大概率是从国内访问被block了。可以挂下代理看看有没有恢复。

[botoai]

是代理问题，可以关掉了。