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-V4-Pro vs Qwen2.5 Instruct 72B

DeepSeekvsAlibaba20 shared benchmarks191 head-to-head
BenchmarkDeepSeek-V4-ProQwen2.5 Instruct 72B
AA Intelligence5310
AA-LCR7020.3
AA-Omniscience-10.6-52.2
Artificial Analysis Coding Index59.411.9
Chinese SimpleQA (C-SimpleQA)77.752.2
Codeforces320624.8
critpt130
GPQA Diamond90.549.1
HLE48.24.2
IFBench76.536.9
livebench73.652.3
LiveCodeBench93.555.5
MMLU-Pro87.571.6
OmniScience Accuracy4317.6
OmniScience Non-Hallucination12.215.3
scicode5026.7
simpleqa57.910.3
SWE-bench Verified80.623.8
Terminal-Bench Hard46.24.5
τ²-Bench Telecom (AA run)96.234.5

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.