0

GLM 5.2

Publisher
Zai
Context
1.0M
Input
Output
Released
16 Jun 2026

Cite

Notes

Only stored in your browser.

Attribution

Benchmark scores
AALiveBenchVals AI
Attribution policy →
Context
1.0M
Input
$1.40
/1M tok
Output
$4.40
/1M tok
Blended
$2.15
#356/ 428/1M · 3:1
Speed
123
#86/ 428tok/s
Cache read
$0.18
0.13x input

Price & speed via Artificial Analysis · ranked across tracked models · caching via OpenRouter

Reported on 22 evals across 9 domains - best 99.1% on τ²-bench (Tau²-bench) · #1 of 333

Reported eval scores

22
τ²-bench (Tau²-bench)99.1
Humanity's Last Exam (HLE)40.1
LiveBench76.7
Vals Index65
Terminal-Bench (Hard)50.8
LiveBench - Coding79.7
IFBench73.3
LiveBench - Math89.8
LiveBench - Language76.2
LiveBench - Data Analysis73.7
LiveBench - Reasoning78.6
LiveBench - Instruction Following62.3
SciCode50.5
GPQA Diamond89.5
Vibe Code Bench v1.164
CorpFin v266.1
Finance Agent (v2)49.7
TaxEval v273.3
ProofBench35
MedScribe83.5