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 Gemma 4 E2B

AnthropicvsGoogle29 shared benchmarks281 head-to-head
BenchmarkClaude Opus 4.5Gemma 4 E2B
AA Agentic Index59.67.4
AA Intelligence41.910
AA-LCR7617
AA-Omniscience14-23.6
aime_202695.137.5
AIME25 no tools9126.3
Artificial Analysis Coding Index47.88.3
Claw-Eval Avg76.653.1
critpt4.60
gdpval47.30
GPQA Diamond8743.4
HLE30.84.8
IFBench5838
LiveCodeBench v684.844
mathvision77.152.4
MathVista80.262.1
MMLU-Pro9060
mmmlu90.867.4
MMMU-Pro7444.6
MMStar73.257.9
OCRBench86.50.7
OmniScience Accuracy46.68.2
OmniScience Non-Hallucination3967.6
RealWorldQA7756.2
scicode5020.9
Tau2 retail88.918.9
Terminal-Bench Hard473
τ²-Bench98.224.5
τ²-Bench Telecom (AA run)89.522.2

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.