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 Opus 4.7 vs GLM-5.2 Full Open Source

AnthropicvsZ.ai41 shared benchmarks2810 head-to-head
BenchmarkClaude Opus 4.7GLM-5.2 Full Open Source
AA Agentic Index64.645.7
AA Intelligence5552.6
AA-LCR75.376.7
AA-Omniscience27.34.4
aime_202695.899.2
arc_agi_19277
ARC-AGI-275.822.8
arena_elo14941471
arena_text_factuality14741460
Artificial Analysis Coding Index73.668.8
coding_arena_elo15571582
critpt1220.9
gdpval58.650.3
GPQA Diamond94.291.2
HiL-Bench41.743.7
HLE54.754.7
HLE (with tools)54.754.7
hmmt_feb_202693.992.5
IFBench58.673.3
itbenchSre46.742.7
livebench_agentic_coding50.751.8
livebench_coding82.179.7
livebench_data_analysis78.373.7
livebench_instruction_following66.762.3
livebench_language77.976.2
livebench_math92.889.8
livebench_reasoning87.278.6
MCP Atlas79.182.6
officeqa_pro80.641.4
OmniScience Accuracy48.924.3
OmniScience Non-Hallucination57.773.7
scicode54.550.5
simplebench61.758.8
simpleqa_verified50.638.1
SWE-bench Pro64.362.1
TauBench V3 - Banking34.634.6
Terminal-Bench 2.183.182.7
Terminal-Bench Hard54.550.8
toolathlon59.348.2
τ²-Bench Telecom (AA run)88.699.1
τ³-Bench28.926.8

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.