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 4.5 vs GLM-5.2 Full Open Source

AnthropicvsZ.ai39 shared benchmarks1325 head-to-head
BenchmarkClaude Opus 4.5GLM-5.2 Full Open Source
AA Agentic Index59.645.7
AA Intelligence41.952.6
AA-LCR7676.7
AA-Omniscience144.4
aime_202695.199.2
arc_agi_18277
ARC-AGI-237.622.8
arena_elo14731471
arena_text_factuality14581460
Artificial Analysis Coding Index47.868.8
coding_arena_elo14951582
critpt4.620.9
gdpval47.350.3
GPQA Diamond8791.2
HLE30.854.7
HLE (with tools)43.454.7
hmmt_feb_202685.392.5
hmmt_nov_202593.394.4
IFBench5873.3
imo_answer_bench8491
livebench_agentic_coding39.751.8
livebench_coding79.779.7
livebench_data_analysis74.473.7
livebench_instruction_following62.562.3
livebench_language81.376.2
livebench_math90.489.8
livebench_reasoning80.178.6
MCP Atlas69.882.6
nl2repo43.248.9
OmniScience Accuracy46.624.3
OmniScience Non-Hallucination3973.7
scicode5050.5
simplebench6258.8
simpleqa_verified41.838.1
SWE-bench Pro57.162.1
Terminal-Bench Hard4750.8
tool_decathlon43.548.2
toolathlon43.548.2
τ²-Bench Telecom (AA run)89.599.1

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.