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 4.7 vs Gemini 2.5 Flash

AnthropicvsGoogle29 shared benchmarks254 head-to-head
BenchmarkClaude Opus 4.7Gemini 2.5 Flash
AA Agentic Index64.618.8
AA Intelligence5517.7
AA-LCR75.365.7
AA-Omniscience27.3-29.8
AIME9156.3
arc_agi_19233.3
ARC-AGI-275.82.5
arena_text_factuality14741437
arena_vision13041214
Artificial Analysis Coding Index73.622.2
critpt121.4
frontiermath_tier_422.94.2
gdpval58.611.9
GPQA Diamond94.282.8
harmbench17.562.6
HLE54.712.1
IFBench58.650.3
MMMU-Pro78.869.1
OmniScience Accuracy48.926.1
OmniScience Non-Hallucination57.724.7
scicode54.535.9
simplebench61.741.2
SWE-bench Verified87.660.4
Terminal-Bench Hard54.513.6
vectara_answer_rate9899
vectara_avg_summary_length149.1101.5
vectara_factual_consistency8892.2
vectara_hallucination_rate127.8
τ²-Bench Telecom (AA run)88.631.6

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.