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 Kimi K2 (Reasoning)

GooglevsMoonshot42 shared benchmarks1624 head-to-head
BenchmarkGemini 2.5 ProKimi K2 (Reasoning)
AA Agentic Index7.247.9
AA Intelligence2733.5
AA-LCR6670.3
AA-Omniscience-14.3-21.4
aider_polyglot82.259.1
AIME 202588.394.5
ArtifactsBench57.754.2
Artificial Analysis Coding Index46.734.8
browsecomp9.960.2
browsecomp_zh32.262.3
critpt2.62.6
FinSearchComp-global42.629.5
frontiermath_tier_44.20
GAIA (text only)60.260.2
gdpval8.524.5
GPQA Diamond84.484.5
HLE22.523.9
HLE (with tools)28.444.9
HMMT 202565.789.4
hmmt_feb_202582.593.3
hmmt_nov_20258089.2
IFBench4968.1
LiveCodeBench82.779.2
MMLU-Pro8684.6
mrcr9344.2
multichallenge53.666.4
OmniScience Accuracy3930.9
OmniScience Non-Hallucination12.625.8
scicode4344.8
simplebench62.439.6
swe_bench_bash53.663.4
SWE-bench Verified67.271.3
Terminal-Bench25.347.1
Terminal-Bench Hard26.531.1
theagentcompany39.330
vectara_answer_rate99.198.6
vectara_avg_summary_length106.459.2
vectara_factual_consistency9382.1
vectara_hallucination_rate717.9
xbench_deepsearch5676
τ²-Bench5474.3
τ²-Bench Telecom (AA run)54.193

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.