BoundaryML/baml · critical

pythonize_checks

Error message

pythonize_checks

What it means

In pythonize_strict (Rust/PyO3), when an LLM response value has checks, the code calls pythonize_checks to build Python validation expressions and unwraps with expect("pythonize_checks"). A panic here means check generation failed — typically the Checked class or model_validate target is missing/misconfigured in the BAML Python runtime.

Solutions

1. Reinstall/upgrade the baml Python package so baml_py.Checked exists
2. Align BAML client codegen version with the engine version
3. Remove or fix custom checks in the BAML schema causing generation failure
4. File a bug with the schema if pythonize_checks panics on valid input

import baml_py
assert hasattr(baml_py, "Checked"), "baml runtime missing Checked support"

def runtime_supports_checks() -> bool:
    import baml_py
    return hasattr(baml_py, "Checked")

When it happens

Trigger: Parsing an LLM response that includes checks when the corresponding Python module lacks the expected Checked infrastructure, or cls_module/model_validate_method is invalid.

Common situations: Mismatched baml-py runtime and Rust client versions; custom class modules that don't provide Checked; corrupted or partial installation of the baml package.

Understand the failure class

Background: "This is a bug, please report it": internal invariant violations, unreachable panics, and SNH errors explained — this error's family across 47 libraries.

Related errors

• get_item
• getattr(Checked)
• PyTuple::new
• ai.Prompt._data must contain baml_builtins2::PromptAst
• ai.Prompt.messages receiver must be an ai.Prompt instance

AI-assisted analysis of BoundaryML/baml@bd85ce9dee (2026-09-12). Data as JSON: /api/errors/36bf93f8801bc87a. Report an issue: GitHub.

Appendix: source

Thrown at engine/language_client_python/src/types/function_results.rs:313

            Ok(instance.into())
        }
        BamlValueWithMeta::Null(_) => Ok(py.None()),
    }?;

    let ResponseValueMeta(_, checks, completion_state, _) = meta;
    if checks.is_empty() && !completion_state.display {
        return Ok(py_value_without_constraints);
    }

    // Import the necessary modules and objects
    let typing = py.import("typing").expect("typing");
    let literal = typing.getattr("Literal").expect("Literal");

    let value_with_possible_checks = if !checks.is_empty() {
        // Generate the Python checks
        let python_checks = pythonize_checks(py, cls_module, &checks, model_validate_method)
            .expect("pythonize_checks");

        // Get the type of the original value
        let value_type = py_value_without_constraints.bind(py).get_type();

        // Collect check names as &str and turn them into a Python tuple
        let check_names: Vec<&str> = checks.iter().map(|check| check.name.as_str()).collect();
        let literal_args = PyTuple::new(py, check_names)?;

        // Call Literal[...] dynamically
        let literal_check_names = literal.get_item(literal_args).expect("get_item");

        let class_checked_type_constructor =
            cls_module.getattr("Checked").expect("getattr(Checked)");

        // Prepare type parameters for Checked[...]
        let type_parameters_tuple =
            PyTuple::new(py, [value_type.as_ref(), &literal_check_names]).expect("PyTuple::new");

View on GitHub (pinned to bd85ce9dee)