This implementation is compliant with the spec allowing the creation of new mnemonics from an entropy from 128 to 256 bits, from mnemonic words or from the the raw entropy+checksum bytes.
Supports English, Spanish, Italian and French, and passes all tests from the Python reference implementation.