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

Claude Sonnet 5 vs DeepSeek-V4-Pro

AnthropicvsDeepSeek35 shared benchmarks296 head-to-head
BenchmarkClaude Sonnet 5DeepSeek-V4-Pro
AA Agentic Index49.763.3
AA Intelligence55.353
AA-LCR7770
AA-Omniscience16.4-10.6
arena_elo14631457
arena_text_factuality14531450
Artificial Analysis Coding Index71.559.4
browsecomp84.783.4
coding_arena_elo15401582
critpt16.913
gdpval54.849
GDPval-AA v216181307
GPQA Diamond91.190.5
HLE57.448.2
livebench_agentic_coding59.442.6
livebench_coding80.770
livebench_data_analysis71.774.5
livebench_instruction_following63.962.4
livebench_language7578.1
livebench_math92.990.7
livebench_reasoning88.782.7
OmniScience Accuracy4043
OmniScience Non-Hallucination60.612.2
scicode53.650
simplebench60.650.9
simpleqa_verified2557.9
SWE-bench Multilingual78.376.2
SWE-bench Pro63.255.4
SWE-bench Verified85.280.6
TauBench V3 - Banking37.330.1
Terminal-Bench 2.080.467.9
Terminal-Bench 2.180.572.1
toolathlon54.351.8
usamo_202679.560.7
τ³-Bench28.225.8

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.