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

DeepSeek-V3 vs Gemini 2.5 Flash

DeepSeekvsGoogle36 shared benchmarks630 head-to-head
BenchmarkDeepSeek-V3Gemini 2.5 Flash
AA Agentic Index1.618.8
AA Intelligence15.417.7
AA-LCR41.365.7
AA-Omniscience-37.6-29.8
aider_polyglot55.161.9
AIME 202459.482.3
AIME 202551.372
AIR-Bench 202440.868.7
anthropic_red_team97.198.8
Artificial Analysis Coding Index2322.2
bbq96.797.7
critpt01.4
gdpval011.9
GPQA Diamond68.482.8
harmbench49.762.6
HLE5.212.1
hmmt_feb_202529.264.2
IFBench4150.3
LiveCodeBench49.676.2
livecodebench_easy83.399.1
livecodebench_hard14.248.3
livecodebench_medium53.582.5
OmniScience Accuracy25.426.1
OmniScience Non-Hallucination23.324.7
scicode35.835.9
simple_safety_tests95.398
simplebench27.241.2
simpleqa27.726.9
SWE-bench Verified4260.4
Terminal-Bench Hard15.213.6
vectara_answer_rate97.599
vectara_avg_summary_length81.7101.5
vectara_factual_consistency93.992.2
vectara_hallucination_rate6.17.8
xstest97.198.8
τ²-Bench Telecom (AA run)47.131.6

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.