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 4.5 vs DeepSeek-R1

AnthropicvsDeepSeek37 shared benchmarks361 head-to-head
BenchmarkClaude Opus 4.5DeepSeek-R1
AA Agentic Index59.63.1
AA Intelligence41.918.6
AA-LCR7656
AA-Omniscience14-31.3
AIME 202592.887.5
arc_agi_18221.2
ARC-AGI-237.61.3
Artificial Analysis Coding Index47.824.6
browsecomp378.9
browsecomp_zh62.435.7
C-Eval92.291.8
critpt4.60.6
Fortress13.674.4
gdpval47.31.5
GPQA Diamond8781
HLE30.817.7
hmmt_feb_202592.976.7
IFBench5839
LiveCodeBench8784.4
longbench_v264.458.3
mmlu_redux95.693.4
MMLU-Pro9085
multichallenge5945
MultiNRC48.627.6
OmniScience Accuracy46.630.7
OmniScience Non-Hallucination3910.5
scicode5035.7
simplebench6240.8
simpleqa_verified41.827.4
SWE-bench Multilingual77.530.5
SWE-bench Verified81.557.6
Terminal-Bench Hard476.1
vectara_answer_rate98.797
vectara_avg_summary_length114.593.5
vectara_factual_consistency89.188.7
vectara_hallucination_rate10.911.3
τ²-Bench Telecom (AA run)89.511.4

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.