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 Sonnet 4.5 vs Gemma 4 E2B

AnthropicvsGoogle27 shared benchmarks261 head-to-head
BenchmarkClaude Sonnet 4.5Gemma 4 E2B
AA Agentic Index50.67.4
AA Intelligence37.410
AA-LCR68.317
AA-Omniscience-0.1-23.6
AIME25 no tools8826.3
Artificial Analysis Coding Index52.18.3
critpt1.10
gdpval40.40
GPQA Diamond83.443.4
HallusionBench59.90.6
HLE19.84.8
IFBench57.338
LiveCodeBench v66444
LiveCodeBenchV6 no tools6454.9
MathVista79.862.1
MMLU-Pro88.260
mmmlu89.167.4
MMMU-Pro68.744.6
OmniScience Accuracy32.98.2
OmniScience Non-Hallucination50.967.6
RealWorldQA70.356.2
scicode4520.9
TauBench V3 - Banking24.54.5
Terminal-Bench 2.155.80.4
Terminal-Bench Hard35.63
τ²-Bench9824.5
τ²-Bench Telecom (AA run)78.122.2

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.