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

Claude Sonnet 4.6 vs GLM-5.2 Full Open Source

AnthropicvsZ.ai40 shared benchmarks1525 head-to-head
BenchmarkClaude Sonnet 4.6GLM-5.2 Full Open Source
AA Agentic Index61.645.7
AA Intelligence48.452.6
AA-LCR7476.7
AA-Omniscience12.24.4
apexAgents2833.7
arc_agi_186.577
ARC-AGI-260.422.8
arena_elo14721471
arena_text_factuality14601460
Artificial Analysis Coding Index6368.8
coding_arena_elo15231582
critpt3.120.9
DeepSWE 1.13044
gdpval54.850.3
GDPval-AA v213811514
GPQA Diamond89.991.2
HLE4954.7
HLE (with tools)46.854.7
IFBench56.673.3
itbenchSre39.842.7
livebench_agentic_coding42.651.8
livebench_coding79.379.7
livebench_data_analysis7873.7
livebench_instruction_following63.262.3
livebench_language76.176.2
livebench_math8789.8
livebench_reasoning84.878.6
MCP Atlas69.582.6
officeqa_pro53.441.4
OmniScience Accuracy40.924.3
OmniScience Non-Hallucination51.673.7
scicode4750.5
simpleqa_verified2938.1
SWE-bench Pro58.162.1
TauBench V3 - Banking34.434.6
Terminal-Bench 2.171.282.7
Terminal-Bench Hard59.150.8
toolathlon49.448.2
τ²-Bench Telecom (AA run)97.999.1
τ³-Bench30.526.8

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.