diff --git a/languedoc.py b/languedoc.py
--- a/languedoc.py
+++ b/languedoc.py
@@ -1,6 +1,8 @@
 import os
 import random
 import itertools
+import json
+import gzip
 
 from shared import preprocess, identify, extract_ngram_freqs, rank_ngram_freqs, Sample
 
@@ -72,6 +74,7 @@ def cross_validate(sample_sets):
 
 DATA_DIR = os.path.join(os.path.dirname(__file__), "data")
 LANG_DIRS = sorted([x.path for x in os.scandir(DATA_DIR)])
+MODEL_PATH = os.path.join(os.path.dirname(__file__), "models.json.gz")
 
 if __name__ == "__main__":
 	samples = []
@@ -89,4 +92,7 @@ if __name__ == "__main__":
 
 				lang_samples.add(text)
 
+	with gzip.open(MODEL_PATH, mode="wt", encoding="utf-8") as f:
+		json.dump([sample_set.create_model().export() for sample_set in samples], f, ensure_ascii=False)
+
 	print(cross_validate(samples))