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

DeepSeek-R1 vs DeepSeek-V4-Pro

DeepSeekvsDeepSeek32 shared benchmarks131 head-to-head
BenchmarkDeepSeek-R1DeepSeek-V4-Pro
AA Agentic Index3.163.3
AA Intelligence18.653
AA-LCR5670
AA-Omniscience-31.3-10.6
Artificial Analysis Coding Index24.659.4
browsecomp8.983.4
Chinese SimpleQA (C-SimpleQA)63.777.7
Codeforces20293206
critpt0.613
gdpval1.549
GPQA Diamond8190.5
HLE17.748.2
IFBench3976.5
LiveCodeBench84.493.5
MMLU-Pro8587.5
OmniScience Accuracy30.743
OmniScience Non-Hallucination10.512.2
scicode35.750
simplebench40.850.9
simpleqa92.357.9
simpleqa_verified27.457.9
SWE-bench Multilingual30.576.2
SWE-bench Verified57.680.6
TauBench V3 - Banking6.430.1
Terminal-Bench 2.119.172.1
Terminal-Bench Hard6.146.2
vectara_answer_rate9797.2
vectara_avg_summary_length93.5153.8
vectara_factual_consistency88.791.4
vectara_hallucination_rate11.38.6
τ²-Bench Telecom (AA run)11.496.2
τ³-Bench6.425.8

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.