ModelBeat
Models/GLM-5.2
All models
Z

Z.ai (Zhipu AI): GLM-5.2

z-ai/glm-5.2

GLM-5.2 is Z.ai (Zhipu AI)’s open-weights AI model, released Jun 16, 2026. It costs $1.40/M input and $4.40/M output tokens, and handles a 1M-token context window. Its strongest use case is agentic & tools, where it ranks in the 78th percentile of the models Model Beat tracks.

GLM 5.2 is a large-scale reasoning model from Z.ai.

Context
1M
Input
$1.40 / 1M
Output
$4.40 / 1M
Cheapest via Relace
$0.97 / $4 / 1M
Released
Jun 16, 2026
Modalities
text→text
Strongest atAgentic & Tools top 22%Coding top 24%

Benchmarks

Scores on standardized evaluations. Higher is better — and rank shows where GLM-5.2 lands among all models tracked on Model Beat.

57.0
Intelligence Index
Epoch AI
57th percentile of tracked models
76.0
Coding Index
Epoch AI
76th percentile of tracked models
78.0
Agentic Index
Epoch AI
78th percentile of tracked models

Reasoning

7 evals
ARC-AGI77.0%

Abstract visual reasoning

ARC-AGI-222.8%

Harder abstract reasoning

GPQA Diamond91.9%

Graduate-level scientific reasoning

Humanity's Last Exam41.1%

Frontier of human expert knowledge

SimpleBench58.8%

Common-sense trick questions

SimpleQA Verified34.2%

Factual accuracy & hallucination

WeirdML70.1%

Novel ML problem-solving

Coding

3 evals
SciCode51.2%

Scientific research coding

SWE-bench Verified78.7%

Real GitHub issue resolution

WebDev Arena1593

Human-rated web development

Math

1 evals
AIME 2024/202586.4%

Olympiad-qualifier math

Agentic & Tools

2 evals
APEX45.2%

Multi-step agentic tasks

τ²-bench99.1%

Tool-agent-user reliability

In the news

28

Changelog

13
  • GLM-5.2: reported APEX score rose from 35.6% to 45.2% (+27%)benchmark · Oct 5, 2026
  • GLM-5.2: cheapest credible provider via Relace now $4 per 1M output tokens (was $3.036, +32%)cheapest provider · Oct 2, 2026
  • GLM-5.2: cheapest credible provider via Alibaba now $0.966 per 1M input tokens (was $0.5, +93%)cheapest provider · Sep 8, 2026
  • GLM-5.2: reported SimpleQA Verified score fell from 38.1% to 34.2% (-10%)benchmark · Aug 28, 2026
  • GLM-5.2: cheapest credible provider via Sail Research now $0.5 per 1M input tokens (was $0.9, -44%)cheapest provider · Aug 13, 2026
  • GLM-5.2: cheapest credible provider via Sail Research now $0.9 per 1M input tokens (was $0.966, -7%)cheapest provider · Aug 8, 2026
  • GLM-5.2: cheapest credible provider via Alibaba now $3.036 per 1M output tokens (was $4.1, -26%)cheapest provider · Jul 31, 2026
  • GLM-5.2: cheapest credible provider via Alibaba now $0.966 per 1M input tokens (was $1.1, -12%)cheapest provider · Jul 31, 2026
  • GLM-5.2: cheapest credible provider via Morph now $4.1 per 1M output tokens (was $3.07, +34%)cheapest provider · Jul 18, 2026
  • GLM-5.2: cheapest credible provider via Morph now $1.1 per 1M input tokens (was $0.97, +13%)cheapest provider · Jul 18, 2026
  • GLM-5.2: cheapest credible provider via Baidu now $3.07 per 1M output tokens (was $1.716, +79%)cheapest provider · Jul 16, 2026
  • GLM-5.2: cheapest credible provider via Baidu now $0.97 per 1M input tokens (was $0.546, +78%)cheapest provider · Jul 16, 2026
  • GLM-5.2 (Z.ai (Zhipu AI)) added to the tracker.tracked · Jun 22, 2026

Compare GLM-5.2 with

Frequently asked questions

What is GLM-5.2?

GLM 5.2 is a large-scale reasoning model from Z.ai.

Who created GLM-5.2?

GLM-5.2 is an AI model developed by Z.ai (Zhipu AI), released Jun 16, 2026.

How much does GLM-5.2 cost?

GLM-5.2's list price from Z.ai (Zhipu AI) is $1.40 per million input tokens and $4.40 per million output tokens. The cheapest credible third-party provider on OpenRouter (Relace) serves it at $0.97/$4 per 1M.

What is GLM-5.2's context window?

GLM-5.2 supports a context window of up to 1M tokens.

How does GLM-5.2 perform on benchmarks?

On standardized evaluations tracked by Epoch AI, GLM-5.2 scores 99.1% on τ²-bench, 91.9% on GPQA Diamond, 78.7% on SWE-bench Verified.

Is GLM-5.2 open source?

GLM-5.2 is released with open weights (Open weights (unrestricted)), so the model can be downloaded and self-hosted.

Benchmark data from Epoch AI (CC BY) and Artificial Analysis; pricing, specs & descriptions from OpenRouter. Data updated Oct 6, 2026.