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-V4-Pro vs Kimi K3

DeepSeekvsMoonshot43 shared benchmarks439 head-to-head
BenchmarkDeepSeek-V4-ProKimi K3
AA Agentic Index63.354.3
AA Intelligence5360
AA-LCR7082.7
AA-Omniscience-10.646
Agents' Last Exam16.528.3
aime_202696.796.7
apexAgents24.341.3
arena_elo14571489
arena_text_factuality14501471
Artificial Analysis Coding Index59.476.2
AutomationBench (Public)12.830.8
browsecomp83.491.2
coding_arena_elo15821674
critpt1323.4
cybergym52.780
DeepSWE12.867.5
DSBench-FullStack41.873.7
DSBench-Hard31.163
gdpval4959.4
GDPval-AA v213071687
GPQA Diamond90.593.5
HLE48.256
HLE (wo / w tools)48.256
hmmt_feb_202695.297
itbenchSre38.347.7
livebench_agentic_coding42.662.2
livebench_coding7081.5
livebench_data_analysis74.578.7
livebench_instruction_following62.471.4
livebench_language78.185.5
livebench_math90.784.4
livebench_reasoning82.790.7
MCP Atlas74.284.2
OmniScience Accuracy4347.6
OmniScience Non-Hallucination12.249.1
Program Bench47.877.8
scicode5058.7
simplebench50.960.7
simpleqa_verified57.942.7
Tau 3 Banking2633.4
TauBench V3 - Banking30.146
Terminal-Bench 2.172.188.3
toolathlon51.873.2

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.