binary-husky/gpt_academic · error · RuntimeError

APIKEY为空,请检查配置文件的{APIKEY}

Error message

APIKEY为空,请检查配置文件的{APIKEY}

What it means

predict_no_ui_long_connection in the OpenAI-compatible template validates that APIKEY (resolved from config/environment) is non-empty before building the request; empty means the model's key variable was never configured, and it raises RuntimeError('APIKEY为空...'). The f-string interpolates the empty key, so the message literally ends with an empty value.

Source

Thrown at request_llms/oai_std_model_template.py:175

        console_silence=False,
    ):
        """
        发送至chatGPT，等待回复，一次性完成，不显示中间过程。但内部用stream的方法避免中途网线被掐。
        inputs：
            是本次问询的输入
        sys_prompt:
            系统静默prompt
        llm_kwargs：
            chatGPT的内部调优参数
        history：
            是之前的对话列表
        observe_window = None：
            用于负责跨越线程传递已经输出的部分，大部分时候仅仅为了fancy的视觉效果，留空即可。observe_window[0]：观测窗。observe_window[1]：看门狗
        """
        from .bridge_all import model_info
        watch_dog_patience = 5  # 看门狗的耐心，设置5秒不准咬人 (咬的也不是人)
        if len(APIKEY) == 0:
            raise RuntimeError(f"APIKEY为空,请检查配置文件的{APIKEY}")
        if inputs == "":
            inputs = "你好👋"


        headers, payload = generate_message(
            input=inputs,
            model=remove_prefix(llm_kwargs["llm_model"]),
            key=APIKEY,
            history=history,
            max_output_token=max_output_token,
            system_prompt=sys_prompt,
            temperature=llm_kwargs["temperature"],
        )

        reasoning = model_info[llm_kwargs['llm_model']].get('enable_reasoning', False)

        retry = 0
        while True:

View on GitHub (pinned to d6bde0fa54)

Solutions

1. Find which key variable the model maps to (config.py APIKEY_LAYOUT / bridge_all model_info) and set it.
2. Set the key via environment variable using the shared_utils/config_loader convention and reload.
3. Restart the app after editing config so the cached config loader picks up the new value.
4. Verify no trailing quotes/whitespace turned the value into an empty string.

Example fix

# config.py
# before
GPT_4_64K_API_KEY = ""

# after
GPT_4_64K_API_KEY = "sk-..."

from toolbox import get_conf
APIKEY = get_conf('GPT_4_API_KEY')  # whichever var the model maps to
if not APIKEY:
    raise ConfigError('GPT_4_API_KEY is empty — set it in config.py or env')

When it happens

Trigger: Using an OpenAI-style model (e.g. via oai_std_model_template) when the corresponding API key config entry (e.g. GPT_4_API_KEY or custom APIKEY layout) is unset/empty in config.py or its environment variable.

Common situations: New deployment where only some keys were filled in; switching LLM_MODEL to a provider whose key variable was never set; docker/env deployment where the key env var was not passed; key set in a different config than the one loaded.

Related errors

• 你提供了错误的API_KEY。 1. 临时解决方案：直接在输入区键入api_key，然后回车提交。 2. 长效解决方
• 你提供了错误的API_KEY。 1. 临时解决方案：直接在输入区键入api_key，然后回车提交。 2. 长效解决方
• 用户代理或助理代理未定义
• AZURE_CFG_ARRAY中配置的模型必须以azure开头
• 模型覆盖参数 '{model_override}' 指向一个暂不支持的模型，请检查配置文件。

AI-assisted analysis of binary-husky/gpt_academic@d6bde0fa54 (2026-08-14). Data as JSON: /api/errors/ec9cb71e3a5e3d43. Report an issue: GitHub.