Warm up runs, then each script is executed 20 times, and we keep the median time. To minimize garbage collection and cross‑benchmark interference, all scenarios are run individually. All benchmark scripts are included so you can reproduce the results locally.
LIST DIFF
Scenario
Superdiff
arr-diff
deep-diff
10k items array
1.84 ms
32.95 ms
4.74 ms
100k items array
17.43 ms
3363.15 ms
50.36 ms
OBJECT DIFF
Scenario
superdiff
deep-object-diff
deep-diff
microdiff
10k flat object keys
2.27 ms
2.44 ms
39.37 ms
2.24 ms
100k flat object keys
29.23 ms
31.86 ms
3784.50 ms
29.51 ms
100k nested nodes
4.25 ms
9.67 ms
16.51 ms
7.26 ms
TEXT DIFF
Scenario
superdiff
diff
10k words
1.38 ms
3.86 ms
100k words
21.68 ms
45.93 ms
10k sentences
2.30 ms
5.61 ms
100k sentences
21.95 ms
62.03 ms
(Superdiff uses its normal accuracy settings to match diff's behavior)
👉 Despite providing a full structural diff with a richer output, Superdiff consistently outperforms or matches the fastest diff libraries. It also scales linearly, even with deeply nested data.
