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 Qwen2.5 Instruct 72B

AnthropicvsAlibaba17 shared benchmarks161 head-to-head
BenchmarkClaude Sonnet 4.6Qwen2.5 Instruct 72B
AA Intelligence48.410
AA-LCR7420.3
AA-Omniscience12.2-52.2
Artificial Analysis Coding Index6311.9
critpt3.10
GPQA Diamond89.949.1
harmbench24.272.8
HLE494.2
IFBench56.636.9
livebench75.552.3
mmmlu89.374.8
OmniScience Accuracy40.917.6
OmniScience Non-Hallucination51.615.3
scicode4726.7
SWE-bench Verified79.623.8
Terminal-Bench Hard59.14.5
τ²-Bench Telecom (AA run)97.934.5

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.