hsliuping/TradingAgents-CN · error · ValueError

不支持的OpenAI兼容提供商: {provider}

Error message

不支持的OpenAI兼容提供商: {provider}

What it means

create_openai_compatible_llm is a factory that only accepts providers registered in the OPENAI_COMPATIBLE_PROVIDERS registry; an unregistered name raises this immediately. The registry maps provider slugs to adapter classes and optional default base URLs, so unknown slugs cannot be dispatched. This is a typo/registration error, not a runtime environment issue.

Source

Thrown at tradingagents/llm_adapters/openai_compatible_base.py:512

            "custom-model": {"context_length": 32768, "supports_function_calling": True}
        }
    }
}


def create_openai_compatible_llm(
    provider: str,
    model: str,
    api_key: Optional[str] = None,
    temperature: float = 0.1,
    max_tokens: Optional[int] = None,
    base_url: Optional[str] = None,
    **kwargs
) -> OpenAICompatibleBase:
    """创建OpenAI兼容LLM实例的统一工厂函数"""
    provider_info = OPENAI_COMPATIBLE_PROVIDERS.get(provider)
    if not provider_info:
        raise ValueError(f"不支持的OpenAI兼容提供商: {provider}")

    adapter_class = provider_info["adapter_class"]

    # 如果调用未提供 base_url，则采用 provider 的默认值（可能为 None）
    if base_url is None:
        base_url = provider_info.get("base_url")

    # 仅当 provider 未内置 base_url（如 custom_openai）时，才将 base_url 传递给适配器，
    # 避免与适配器内部的 super().__init__(..., base_url=...) 冲突导致 "multiple values" 错误。
    init_kwargs = dict(
        model=model,
        api_key=api_key,
        temperature=temperature,
        max_tokens=max_tokens,
        **kwargs,
    )
    if provider_info.get("base_url") is None and base_url:
        init_kwargs["base_url"] = base_url

View on GitHub (pinned to 74783e8817)

Solutions

1. Check valid slugs via list(OPENAI_COMPATIBLE_PROVIDERS) in openai_compatible_base.py and use an exact match (lowercase)
2. Register your custom adapter: OPENAI_COMPATIBLE_PROVIDERS['myprov'] = {'adapter_class': MyProvAdapter, 'base_url': '...'}
3. Normalize/validate provider strings from config or UI before calling the factory

Example fix

# before
llm = create_openai_compatible_llm(provider="DeepSeek", model="deepseek-chat")  # case mismatch

# after
llm = create_openai_compatible_llm(provider="deepseek", model="deepseek-chat")  # exact registry slug

from tradingagents.llm_adapters.openai_compatible_base import OPENAI_COMPATIBLE_PROVIDERS
provider = provider.strip().lower()
if provider not in OPENAI_COMPATIBLE_PROVIDERS:
    raise SystemExit(f"Unknown provider '{provider}'. Valid: {sorted(OPENAI_COMPATIBLE_PROVIDERS)}")

def is_supported_provider(provider: str) -> bool:
    from tradingagents.llm_adapters.openai_compatible_base import OPENAI_COMPATIBLE_PROVIDERS
    return provider.strip().lower() in OPENAI_COMPATIBLE_PROVIDERS

try:
    llm = create_openai_compatible_llm(provider=p, model=m)
except ValueError as e:
    if "不支持的OpenAI兼容提供商" in str(e):
        raise SystemExit(f"Unknown provider '{p}'; check registry") from e
    raise

When it happens

Trigger: Calling create_openai_compatible_llm(provider="openai") or any slug not in OPENAI_COMPATIBLE_PROVIDERS (e.g. 'gpt', 'azure', 'googl'), or a custom provider that was never registered.

Common situations: Provider string from user input or config passed unsanitized; provider slug renamed between versions; new adapter written but not added to the registry dict; case mismatch ('DeepSeek' vs 'deepseek').

Related errors

• Indicator {indicator} is not supported. Please choose from:
• REDDIT FETCHING ERROR: max limit is less than the number of
• Trading Agents Graph Setup Error: no analysts selected!
• QIANFAN_API_KEY格式错误，应为: bce-v3/ALTAK-xxx/xxx
• Unsupported LLM provider: {provider}

AI-assisted analysis of hsliuping/TradingAgents-CN@74783e8817 (2026-08-28). Data as JSON: /api/errors/9b16eed1da732502. Report an issue: GitHub.