VECTOR WIREAI INTELLIGENCE
NVDA$1,847+3.2%MSFT$512+1.1%GOOGL$199-0.4%META$728+2.7%AMD$184-1.2%TSM$212+0.6%PLTR$98+4.1%AI IDX4,821+1.9%
PKT
SEEDRefresh Models Deals Regulatory Sources

Nemotron 3 Super 120B A12B vs Qwen3.5 35B A3B

NVIDIAvsAlibaba44 shared benchmarks2420 head-to-head
BenchmarkNemotron 3 Super 120B A12BQwen3.5 35B A3B
AA Agentic Index8.844.1
AA Intelligence25.729.9
AA-LCR60.368.3
AA-Omniscience-41.5-48.1
AGIEval-En77.970.5
aime_202691.793.3
arc_challenge96.195.4
Artificial Analysis Coding Index37.737
browsecomp31.361
critpt3.10.9
gdpval9.920.3
Global-MMLU-Lite (5-shot) Average85.780.9
GPQA Diamond82.784.5
GSM8K90.790.1
hellaswag8985.6
HLE22.847.4
HMMT 202594.789.2
hmmt_feb_202685.281.8
humaneval80.566.5
IFBench72.672.5
itbenchSre1.121.5
MBPP81.770.8
Minerva Math84.859.7
mmlu8681.1
mmlu_prox79.481
MMLU-Pro83.885.3
multichallenge55.260
OmniScience Accuracy24.320.1
OmniScience Non-Hallucination1314.6
OpenBookQA48.644.2
PIQA (acc-norm)85.582.5
PIQA (acc)83.982.3
RULER 1M93.956.4
RULER 256K96.782.4
scicode4237.7
SWE-bench Multilingual45.860.3
SWE-bench Verified53.770
TauBench V3 - Banking10.34.9
Terminal-Bench 2.03140.5
Terminal-Bench 2.138.640.8
Terminal-Bench Hard28.826.5
winogrande78.979.2
τ²-Bench Telecom (AA run)67.889.2
τ³-Bench10.168.9

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.