Which is better, NVIDIA-Nemotron-3.5-Lightning-30B-A3B-Base-BF16 or Qwen3.5 35B A3B?

Refresh Models —Deals —Regulatory —Sources —

Questions this page answers

Which is better, NVIDIA-Nemotron-3.5-Lightning-30B-A3B-Base-BF16 or Qwen3.5 35B A3B?: Across 16 shared benchmarks, NVIDIA-Nemotron-3.5-Lightning-30B-A3B-Base-BF16 scores higher on 9 and Qwen3.5 35B A3B on 7. The widest gap is Minerva Math, where NVIDIA-Nemotron-3.5-Lightning-30B-A3B-Base-BF16 scores 82.8 against 59.7.

NVIDIA-Nemotron-3.5-Lightning-30B-A3B-Base-BF16 vs Qwen3.5 35B A3B

Across 16 shared benchmarks, NVIDIA-Nemotron-3.5-Lightning-30B-A3B-Base-BF16 scores higher on 9 and Qwen3.5 35B A3B on 7. The widest gap is Minerva Math, where NVIDIA-Nemotron-3.5-Lightning-30B-A3B-Base-BF16 scores 82.8 against 59.7.

NVIDIAvsAlibaba16 shared benchmarks9–7 head-to-head

Benchmark	NVIDIA-Nemotron-3.5-Lightning-30B-A3B-Base-BF16	Qwen3.5 35B A3B
AGIEval-En	70	70.5
arc_challenge	92.7	95.4
Global-MMLU-Lite (5-shot) Average	75.5	80.9
GSM8K	91.3	90.1
hellaswag	85.5	85.6
humaneval	77.4	66.5
MBPP	78.6	70.8
Minerva Math	82.8	59.7
mmlu	78.6	81.1
MMLU-Pro	67.9	85.3
OpenBookQA	47.6	44.2
PIQA (acc-norm)	85.2	82.5
PIQA (acc)	83.3	82.3
RULER 1M	69.6	56.4
RULER 256K	76.9	82.4
winogrande	80	79.2

Best tracked score per model per benchmark (default configuration; source-attributed). ↓ marks lower-is-better metrics. Open either model for its full surface, provenance and pricing.