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

Kimi K2 (Reasoning) vs Qwen3.6 27B

MoonshotvsAlibaba25 shared benchmarks520 head-to-head
BenchmarkKimi K2 (Reasoning)Qwen3.6 27B
AA Agentic Index47.927.5
AA Intelligence33.537.7
AA-LCR70.373.3
AA-Omniscience-21.4-20
Artificial Analysis Coding Index34.853.7
critpt2.61.1
gdpval24.532
GPQA Diamond84.587.8
HLE23.924
HMMT 202589.490.7
hmmt_feb_202593.393.8
hmmt_nov_202589.290.7
IFBench68.169.1
imo_answer_bench78.680.8
LiveCodeBench v683.183.9
mmlu_redux94.493.5
MMLU-Pro84.686.2
OmniScience Accuracy30.919.6
OmniScience Non-Hallucination25.850.7
scicode44.839.8
SWE-bench Multilingual61.171.3
SWE-bench Verified71.377.2
Terminal-Bench 2.035.759.3
Terminal-Bench Hard31.134.8
τ²-Bench Telecom (AA run)9394.2

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.