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

Gemini 2.5 Pro vs Gemini 3 Flash Preview

GooglevsGoogle40 shared benchmarks732 head-to-head
BenchmarkGemini 2.5 ProGemini 3 Flash Preview
AA Agentic Index7.249.7
AA Intelligence2738.7
AA-LCR6673
AA-Omniscience-14.310.1
AIME 202588.399.7
arc_agi_14184.7
ARC-AGI-24.933.6
arena_elo14461473
arena_text_factuality14621452
arena_vision12461260
Artificial Analysis Coding Index46.742.6
coding_arena_elo12241387
critpt2.68.6
frontiermath_tier_44.24.2
gdpval8.535.2
GPQA Diamond84.490.4
harmbench65.470
HLE22.543.5
hmmt_feb_202582.597.5
hmmt_nov_20258093.3
IFBench4978
matharena_visual_math_overall77.285.8
MMMU-Pro74.981.2
OmniScience Accuracy3953.4
OmniScience Non-Hallucination12.67.6
scicode4350.6
simplebench62.461.1
simpleqa50.868.7
simpleqa_verified5667.4
swe_bench_bash53.675.8
SWE-bench Verified67.278
TauBench V3 - Banking9.720.8
Terminal-Bench 2.128.558
Terminal-Bench Hard26.538.6
vectara_answer_rate99.199.8
vectara_avg_summary_length106.490.2
vectara_factual_consistency9386.5
vectara_hallucination_rate713.5
τ²-Bench Telecom (AA run)54.180.4
τ³-Bench9.317.5

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.