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-Flash

AnthropicvsDeepSeek34 shared benchmarks259 head-to-head
BenchmarkClaude Sonnet 5DeepSeek-V4-Flash
AA Agentic Index49.761.3
AA Intelligence55.351.8
AA-LCR7774.3
AA-Omniscience16.4-14.3
arena_elo14631439
arena_text_factuality14531434
Artificial Analysis Coding Index71.569.1
browsecomp84.773.2
coding_arena_elo15401431
critpt16.916.6
gdpval54.852.9
GDPval-AA v216181189
GPQA Diamond91.190.8
HLE57.438.6
livebench_agentic_coding59.446.8
livebench_coding80.775
livebench_data_analysis71.779.3
livebench_instruction_following63.965.5
livebench_language7579.2
livebench_math92.986.8
livebench_reasoning88.786.6
OmniScience Accuracy4040.4
OmniScience Non-Hallucination60.615.6
scicode53.649.9
simplebench60.646.3
simpleqa_verified2545
SWE-bench Multilingual78.373.3
SWE-bench Pro63.252.6
SWE-bench Verified85.279
TauBench V3 - Banking37.339.4
Terminal-Bench 2.080.456.9
Terminal-Bench 2.180.582.7
toolathlon54.370.3
τ³-Bench28.222.9

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.