VECTOR WIREAI INTELLIGENCE
PKT
Refresh Models Deals Regulatory Sources

Questions this page answers

Which model leads MMBench EN-DEV-v1.1?
Across 7 models scored on MMBench EN-DEV-v1.1, Qwen3.6 35B A3B leads at 92.8, ahead of Qwen3.5 27B at 92.6. The median tracked score is 91.5, and the field spans 88.3 to 92.8.

MMBench EN-DEV-v1.1

Across 7 models scored on MMBench EN-DEV-v1.1, Qwen3.6 35B A3B leads at 92.8, ahead of Qwen3.5 27B at 92.6. The median tracked score is 91.5, and the field spans 88.3 to 92.8.

7 models tracked
Data as of August 24, 2026
#ModelVendorBest scoreRunsLast seen
1Qwen3.6 35B A3BAlibaba92.812026-08-24
2Qwen3.5 27BAlibaba92.622026-08-24
3Qwen3.6 27BAlibaba92.312026-08-24
4Qwen3.5 35B A3BAlibaba91.512026-08-24
5Gemma 4 31BGoogle90.922026-08-24
6Gemma 4 26B A4BGoogle8912026-08-24
7Claude Sonnet 4.5Anthropic88.312026-08-24

Best tracked score per model (default configuration; source-attributed and verification-tiered). Open a model for its full benchmark surface, provenance and pricing.