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 Opus 4.8 vs DeepSeek-R1

AnthropicvsDeepSeek27 shared benchmarks261 head-to-head
BenchmarkClaude Opus 4.8DeepSeek-R1
AA Agentic Index49.43.1
AA Intelligence57.318.6
AA-LCR7356
AA-Omniscience28.8-31.3
arc_agi_192.521.2
ARC-AGI-272.11.3
Artificial Analysis Coding Index74.324.6
browsecomp88.58.9
critpt20.90.6
Fortress18.274.4
gdpval54.21.5
GPQA Diamond93.681
HLE57.917.7
IFBench62.239
longbench_v269.158.3
OmniScience Accuracy48.830.7
OmniScience Non-Hallucination60.710.5
scicode53.535.7
simplebench64.840.8
simpleqa_verified39.527.4
SWE-bench Multilingual84.430.5
SWE-bench Verified88.657.6
TauBench V3 - Banking34.26.4
Terminal-Bench 2.18519.1
Terminal-Bench Hard58.36.1
τ²-Bench Telecom (AA run)94.411.4
τ³-Bench27.66.4

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.