← All comparisons

Claude 4.5 Haiku (Reasoning) vs GLM-4.6 (Non-reasoning)

Anthropic vs Z AI — side-by-side benchmark comparison

Claude 4.5 Haiku (Reasoning)GLM-4.6 (Non-reasoning)
Intelligence Index37.130.2
Coding Index32.630.2
Math Index83.744.3
Output speed (tok/s)142.253.2
Blended price ($/1M)$2.19$1.00
Time to first token (s)10.48s1.84s
aime
aime 2583.7%44.3%
artificial analysis coding index32.6030.20
artificial analysis intelligence index37.1030.20
artificial analysis math index83.7044.30
gpqa67.2%63.2%
hle9.7%5.2%
ifbench54.3%36.7%
lcr70.3%26.3%
livecodebench61.5%56.1%
math 500
mmlu pro76.0%78.4%
scicode43.3%33.1%
tau254.7%76.9%
terminalbench hard27.3%28.8%

Benchmark data from Artificial Analysis.