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 Qwen3.7 Plus Preview

DeepSeekvsAlibaba32 shared benchmarks2012 head-to-head
BenchmarkDeepSeek-V4-ProQwen3.7 Plus Preview
AA Agentic Index63.320.8
AA Intelligence5339.4
AA-LCR7069
AA-Omniscience-10.62.4
Agents' Last Exam16.533.6
Apex (Pass@1)38.322.7
apexAgents24.322.4
arena_elo14571458
arena_text_factuality14501454
Artificial Analysis Coding Index59.455.9
critpt139.1
gdpval4922.2
GPQA Diamond90.590.3
HLE48.235.6
hmmt_feb_202695.292.9
IFBench76.579.1
imo_answer_bench89.886
MCP Atlas74.273.2
MMLU-Pro87.588.5
nl2repo38.541.1
OmniScience Accuracy4322.5
OmniScience Non-Hallucination12.274.5
scicode5051.3
SWE-bench Multilingual76.275.8
SWE-bench Pro55.457.6
SWE-bench Verified80.677.7
TauBench V3 - Banking30.117.9
Terminal-Bench 2.067.970.3
Terminal-Bench 2.172.161
Terminal-Bench Hard46.247
τ²-Bench Telecom (AA run)96.293
τ³-Bench25.817.9

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.