What causes "Invalid regex: {e}" in invoke-ai/InvokeAI?

Fix the regex syntax reported in the message (e.g. close the group, escape the bracket)

invoke-ai/InvokeAI · error · ValueError

Invalid regex

Error message

Invalid regex: {e}

What it means

validate_url_regex is a pydantic field validator that compiles the configured URL-filter regex with re.compile(); if re.error is raised the ValueError 'Invalid regex: {e}' surfaces. This guarantees the app never stores an un-compilable regex used for URL filtering.

Solutions

Fix the regex syntax reported in the message (e.g. close the group, escape the bracket)
Validate the pattern quickly in Python: re.compile(your_pattern) before putting it in config
Prefer raw strings in YAML/ENV to preserve backslashes (use single quotes in YAML)

Example fix

// before (invokeai.yaml)
url_regex: '.*\.(png|jpg'   # unclosed group
// after
url_regex: '.*\.(png|jpg)$'

Defensive patterns

Strategy: validation

Validate before calling

import re
try:
    re.compile(cfg.url_regex)
except re.error as e:
    raise ValueError(f"fix url_regex in config: {e}")

Try / catch

try:
    InvokeAIAppConfig(url_regex=pattern)
except ValueError as e:
    if str(e).startswith('Invalid regex'):
        print('Fix the regex in invokeai.yaml:', e)

Prevention

Test regexes with re.compile before adding to config
Use YAML single quotes to preserve backslashes
Avoid porting JS-specific regex syntax into Python

When it happens

Trigger: Setting the url_regex config field (env var, config file, or CLI) to a syntactically invalid regex such as 'foo(' , '[unclosed' , or '*quantifier'.

Common situations: Hand-edited invokeai.yaml with an escaping mistake (e.g. '\.' vs '.') ; quoting issues where backslashes are eaten by YAML/env parsing; porting a JavaScript regex with unsupported syntax into Python.

Related errors

AI-assisted analysis of invoke-ai/InvokeAI@0b6a024f2f (2026-08-29). Data as JSON: /api/errors/b4f46aff88ffbb71. Report an issue: GitHub.

Appendix: source

Thrown at invokeai/app/services/config/config_default.py:62

    "external_openai_api_key",
    "external_openai_base_url",
    "external_seedream_api_key",
    "external_seedream_base_url",
)


class URLRegexTokenPair(BaseModel):
    url_regex: str = Field(description="Regular expression to match against the URL")
    token: str = Field(description="Token to use when the URL matches the regex")

    @field_validator("url_regex")
    @classmethod
    def validate_url_regex(cls, v: str) -> str:
        """Validate that the value is a valid regex."""
        try:
            re.compile(v)
        except re.error as e:
            raise ValueError(f"Invalid regex: {e}")
        return v


class InvokeAIAppConfig(BaseSettings):
    """Invoke's global app configuration.

    Typically, you won't need to interact with this class directly. Instead, use the `get_config` function from `invokeai.app.services.config` to get a singleton config object.

    Attributes:
        host: IP address to bind to. Use `0.0.0.0` to serve to your local network.
        port: Port to bind to.
        allow_origins: Allowed CORS origins.
        allow_credentials: Allow CORS credentials.
        allow_methods: Methods allowed for CORS.
        allow_headers: Headers allowed for CORS.
        ssl_certfile: SSL certificate file for HTTPS. See https://www.uvicorn.dev/settings/#https.
        ssl_keyfile: SSL key file for HTTPS. See https://www.uvicorn.dev/settings/#https.
        log_tokenization: Enable logging of parsed prompt tokens.

View on GitHub (pinned to 0b6a024f2f)