Z.ai logo

GLM-5.2

Z.ai · Released Jun 16, 2026

Open SourcecodingreasoningagentsCompare this model

$0.9

Input price / MTok

#11 of 27

$2.86

Output price / MTok

#9 of 27

1.0M

Context window

#13 of 28

62.1

SWE-bench Pro

#4 of 9

91.2

GPQA Diamond

#4 of 19

99.2

AIME 2026

#1 of 1

Summary

GLM-5.2 is a open-source language model from Z.ai, released on Jun 16, 2026.

At $0.9 input and $2.86 output per million tokens, it ranks #11 of 27 for input price, placing it mid-range on input cost.

Its context window of 1.0M tokens ranks #13 of 28 for context depth in this dataset.

It targets coding, reasoning, agents workloads.

744B total / 40B active MoE; 1M-token context with IndexShare attention (2.9x per-token FLOPs reduction at full context). MIT license. Two reasoning modes: High and Max.

Benchmarks

BenchmarkScoreMax%RankSource
SWE-bench Pro62.110062.1%#4/9Z.ai (self-reported in official Hugging ...
GPQA Diamond91.210091.2%#4/19Z.ai (self-reported in official Hugging ...
AIME 202699.210099.2%only entryZ.ai (self-reported in official Hugging ...

Similar Models

Compare all 6 models

Models with the most shared benchmark coverage and closest scores to GLM-5.2.

GLM-5.2
TencentTencent Hy3
AnthropicClaude Sonnet 4.6
OpenAIGPT-5.5
AnthropicClaude Opus 4.8
AnthropicClaude Sonnet 5
SWE-bench Pro#4 of 9
Professional software engineering tasks
GPQA Diamond#4 of 19
Graduate-level science reasoning

Compare GLM-5.2 side by side

Pre-filled with 5 similar models

Open comparison