Codon 是一個能將符合規範的 Python 程式碼轉換成機械碼的編譯器（compiler）。工具本身是可擴充的，所以可以逐步延伸工具的容納範圍。比較一下 Numba：
| Feature | Codon | Numba |
|---|---|---|
| Compilation Approach | Transpiles Python code into native machine code via a standalone compiler (ahead-of-time). | JIT compilation at runtime, optimizing functions as needed. |
| Language Features | Supports a statically-typed subset of Python; limited to certain Python features. | Supports a wide range of Python features, though with some limitations for advanced features. |
| Performance | Provides high performance by compiling to native machine code. | Offers significant speedups, especially for numerical computations, but is bounded by Python runtime limitations. |
| Interoperability | Requires adaptation to static typing; limited support for Python libraries. | Works within the Python ecosystem, supporting many libraries with minimal code changes. |
| Use Cases | Best for scenarios requiring maximum performance and where code can be adapted to static typing. | Ideal for incrementally optimizing existing Python codebases, especially in scientific computing. |
速度上可以有感提升？不過要確定程式符合其子集語法的要求。
from time import time
def fib(n):
return n if n < 2 else fib(n - 1) + fib(n - 2)
t0 = time()
ans = fib(40)
t1 = time()
print(f'Computed fib(40) = {ans} in {t1 - t0} seconds.')實際執行速度
$ python3 fib.py
Computed fib(40) = 102334155 in 17.979357957839966 seconds.
$ codon run -release fib.py
Computed fib(40) = 102334155 in 0.275645 seconds.