They publish a benchmark for anyone interested: https://marisa-trie.readthedocs.io/en/latest/benchmarks.html https://marisa-trie.readthedocs.io/en/latest/benchm

They publish a benchmark for anyone interested: https://marisa-trie.readthedocs.io/en/latest/benchmarks.html https://marisa-trie.readthedocs.io/en/latest/benchmarks.html Summary for storing a list of 3M Russian words: - ~60x less memory - ~5-10x slower compared to hashmap