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 GPT-5.4

DeepSeekvsOpenAI31 shared benchmarks130 head-to-head
BenchmarkDeepSeek-R1GPT-5.4
AA Agentic Index3.158.2
AA Intelligence18.653.1
AA-LCR5677.7
AA-Omniscience-31.35.8
arc_agi_121.293.7
ARC-AGI-21.374
Artificial Analysis Coding Index24.671.1
browsecomp8.982.7
Chinese SimpleQA (C-SimpleQA)63.776.8
Codeforces20293168
critpt0.623.4
gdpval1.550.1
GPQA Diamond8193
HLE17.743.7
IFBench3973.9
MMLU-Pro8587.5
MultiNRC27.658.3
OmniScience Accuracy30.750.9
OmniScience Non-Hallucination10.517.4
scicode35.756.6
simpleqa_verified27.445.3
SWE-bench Multilingual30.571.7
TauBench V3 - Banking6.439.6
Terminal-Bench 2.119.178.3
Terminal-Bench Hard6.157.6
vectara_answer_rate9799.9
vectara_avg_summary_length93.581.7
vectara_factual_consistency88.793
vectara_hallucination_rate11.37
τ²-Bench Telecom (AA run)11.487.1
τ³-Bench6.472.9

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.