VECTOR WIREAI INTELLIGENCE
PKT
Refresh Models Deals Regulatory Sources

Questions this page answers

Which model leads Global-MMLU-Lite (5-shot) Average?
Across 5 models scored on Global-MMLU-Lite (5-shot) Average, Nemotron 3 Super 120B A12B leads at 85.7, ahead of Qwen3.5 35B A3B at 80.9. The median tracked score is 75.5, and the field spans 74.6 to 85.7.

Global-MMLU-Lite (5-shot) Average

Across 5 models scored on Global-MMLU-Lite (5-shot) Average, Nemotron 3 Super 120B A12B leads at 85.7, ahead of Qwen3.5 35B A3B at 80.9. The median tracked score is 75.5, and the field spans 74.6 to 85.7.

5 models tracked
Data as of August 24, 2026
#ModelVendorBest scoreRunsLast seen
1Nemotron 3 Super 120B A12BNVIDIA85.712026-08-24
2Qwen3.5 35B A3BAlibaba80.912026-08-24
3NVIDIA-Nemotron-3.5-Lightning-30B-A3B-Base-BF16NVIDIA75.512026-08-24
4Gemma 4 26B A4BGoogle74.812026-08-24
5NVIDIA Nemotron 3 Nano 30B A3BNVIDIA74.612026-08-24

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