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 Gemini 2.5 Pro

AnthropicvsGoogle36 shared benchmarks342 head-to-head
BenchmarkClaude Opus 4.8Gemini 2.5 Pro
AA Agentic Index49.47.2
AA Intelligence57.327
AA-LCR7366
AA-Omniscience28.8-14.3
arc_agi_192.541
ARC-AGI-272.14.9
arena_elo14821446
arena_text_factuality14641462
arena_vision12801246
Artificial Analysis Coding Index74.346.7
browsecomp88.59.9
coding_arena_elo15391224
critpt20.92.6
Fortress18.254.9
frontiermath_tier_431.34.2
gdpval54.28.5
GPQA Diamond93.684.4
HLE57.922.5
HLE (with tools)57.928.4
hmmt_nov_202596.580
IFBench62.249
matharena_visual_math_overall81.677.2
mathvision86.773.3
MMMU-Pro78.974.9
OmniScience Accuracy48.839
OmniScience Non-Hallucination60.712.6
scicode53.543
simplebench64.862.4
simpleqa_verified39.556
SWE-bench Verified88.667.2
TauBench V3 - Banking34.29.7
Terminal-Bench 2.18528.5
Terminal-Bench Hard58.326.5
Video-MME8684.8
τ²-Bench Telecom (AA run)94.454.1
τ³-Bench27.69.3

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.