how should i parallel run py.run?

[Budome]

async fn run_python_code(code: Arc, i: usize) -> PyResult<Py> {
Python::attach(|py| {
assert!(py.version_info() >= (3, 13));
let locals = PyDict::new(py);
py.run(code.as_c_str(), Some(&locals), Some(&locals))?;
Ok(locals.into())
})
}

code is
let input_fib = r#"
def fib(x):
if x == 0:
return 0
elif x == 1:
return 1
return fib(x-1) + fib(x-2)
_result = fib(25)
"#;

Whether running individually or in parallel, the total time remains the same. I think it's because of GIL, how can I make it run in parallel?

[davidhewitt]

To do parallelism at the Python level, your best option is probably to consider using Python 3.14's free-threaded build, see https://pyo3.rs/v0.26.0/free-threading.html

(note 3.14 is in the final RC at present)