Codesota · Models · Qwen3 MaxAlibaba Cloud8 results · 2 benchmarks
Model card

Qwen3 Max.

Alibaba Cloudopen-source
§ 01 · Benchmarks

Every benchmark Qwen3 Max has a recorded score for.

#BenchmarkArea · TaskMetricValueRankDateSource
01SWE-bench VerifiedAgentic AI · SWE-benchresolve-rate69.6%#40/81source ↗
02PLCCNatural Language Processing · Polish Cultural Competencygeography75.0%#76/165source ↗
03PLCCNatural Language Processing · Polish Cultural Competencyhistory74.0%#79/165source ↗
04PLCCNatural Language Processing · Polish Cultural Competencyvocabulary54.0%#84/165source ↗
05PLCCNatural Language Processing · Polish Cultural Competencygrammar58.0%#86/165source ↗
06PLCCNatural Language Processing · Polish Cultural Competencyaverage61.3%#86/165source ↗
07PLCCNatural Language Processing · Polish Cultural Competencyart-and-entertainment50.0%#88/165source ↗
08PLCCNatural Language Processing · Polish Cultural Competencyculture-and-tradition57.0%#94/165source ↗
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.
§ 02 · Strengths by area

Where Qwen3 Max actually performs.

Agentic AI
1
benchmark
avg rank #40.0
Natural Language Processing
1
benchmark
avg rank #84.7
§ 04 · Related models

Other Alibaba Cloud models scored on Codesota.

Qwen3.6 Plus
3 results
Qwen3-Coder 480B A35B
2 results
Qwen3.5-397B-A17B
2 results
Qwen3.5-122B-A10B
1 result
Qwen3.5-27B
1 result
Qwen3.5-35B-A3B
1 result
§ 05 · Sources & freshness

Where these numbers come from.

sdadas/PLCC
7
results
editorial
1
result
8 of 8 rows marked verified.