Codesota · Models · Qwen3.5-4BAlibaba7 results · 1 benchmarks
Model card

Qwen3.5-4B.

Alibabaopen-source
§ 02 · Benchmarks

Every benchmark Qwen3.5-4B has a recorded score for.

#BenchmarkArea · TaskMetricValueRankDateSource
01PLCCNatural Language Processing · Polish Cultural Competencygrammar45.0%#130/165—source ↗
02PLCCNatural Language Processing · Polish Cultural Competencyvocabulary34.0%#139/165—source ↗
03PLCCNatural Language Processing · Polish Cultural Competencyculture-and-tradition24.0%#150/165—source ↗
04PLCCNatural Language Processing · Polish Cultural Competencyaverage29.7%#151/165—source ↗
05PLCCNatural Language Processing · Polish Cultural Competencyhistory36.0%#155/165—source ↗
06PLCCNatural Language Processing · Polish Cultural Competencygeography27.0%#156/165—source ↗
07PLCCNatural Language Processing · Polish Cultural Competencyart-and-entertainment12.0%#161/165—source ↗
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 03 · Strengths by area

Where Qwen3.5-4B actually performs.

Natural Language Processing
1
benchmark
avg rank #148.9
§ 05 · Related models

Other Alibaba models scored on Codesota.

Qwen3-235B-A22B
235B (22B active) params · 21 results
Qwen3.5-397B-A17B
20 results
Qwen2-VL 72B
18 results
Qwen3.5-122B-A10B
16 results
Qwen3.5-27B
16 results
Qwen3.5-35B-A3B
16 results
Qwen2-VL 7B
7B params · 11 results
Qwen2.5-14B-Instruct
9 results
§ 06 · Sources & freshness

Where these numbers come from.

sdadas/PLCC
7
results
7 of 7 rows marked verified.