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 K2.5 vs Qwen3.6 35B A3B

MoonshotvsAlibaba45 shared benchmarks395 head-to-head
BenchmarkKimi K2.5Qwen3.6 35B A3B
AA Agentic Index52.821.6
AA Intelligence3632.1
AA-LCR7366.7
AA-Omniscience-7.3-22.2
aime_202695.892.7
Artificial Analysis Coding Index46.841.9
browsecomp74.948.7
charxiv_rq78.778
critpt3.10.3
gdpval38.327.8
GDPval-AA v210091015
GPQA Diamond87.986
HLE50.222.2
HMMT 202595.489.1
hmmt_feb_202595.490.7
hmmt_feb_202687.183.6
hmmt_nov_202591.189.1
IFBench70.264.4
imo_answer_bench81.878.9
LiveCodeBench v68580.4
LVBench75.971.4
MathVista90.186.4
MCP Atlas6462.8
mcpmark29.537
MMLU-Pro87.185.6
MMMU-Pro78.575.3
nl2repo3229.4
OmniDocBench 1.588.889.9
OmniScience Accuracy35.218.8
OmniScience Non-Hallucination5049.5
scicode4935.8
SimpleVQA71.258.9
SWE-bench Multilingual7367.2
SWE-bench Pro53.849.5
SWE-bench Verified76.873.4
TauBench V3 - Banking14.29.3
Terminal-Bench 2.050.851.5
Terminal-Bench 2.145.744.9
Terminal-Bench Hard34.834.8
tool_decathlon27.826.9
toolathlon27.826.9
VideoMMMU86.683.7
WideSearch7960.1
τ²-Bench Telecom (AA run)95.995.3
τ³-Bench6667.2

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.