How much does GLM 5.2 cost?

GLM 5.2 costs $1.00 per million input tokens and $4.00 per million output tokens. For a typical conversation (~2,000 tokens), that's approximately $0.006 per message.

How does GLM 5.2 compare to GPT-5.6 Luna?

GLM 5.2 has an average score of 64.7 while GPT-5.6 Luna scores 64.8. GPT-5.6 Luna slightly outperforms GLM 5.2 overall. GLM 5.2 costs $1.00/1M input vs GPT-5.6 Luna at $0.20/1M input. See full comparison →

Home/Models/GLM 5.2

GLM 5.2

Name: GLM 5.2 · benchmarks, pricing and providers
Creator: BenchGecko

by z-ai · Released Jun 2026

Open Source1M Context

64.7

avg score

Rank #60

Compare

Better than 81% of all models

Context

1.0M tokens (~524 books)

Input $/1M

$1.00

Output $/1M

$4.00

Type

text

License

Open Source

Benchmarks

33 tested

Data as of 2026-10-05

About

GLM 5.2 is a large-scale reasoning model from Z.ai. It supports text input and output with a 1M-token context window, and is suited for long-horizon agent workflows, project-level software engineering,...

Tested on 33 benchmarks · BenchGecko score 64.7. Top scores: Chatbot Arena Elo — Coding (1593.3%), Chatbot Arena Elo — Overall (1471.0%), LiveBench — Mathematics (89.8%).

Looking for similar performance at lower cost?
GPT-5.6 Luna scores 64.8 (100% as good) at $0.20/1M input · 80% cheaper

Capabilities

coding

61.7

#41 globally

reasoning

60.5

#58 globally

math

66.2

#57 globally

knowledge

49.6

#129 globally

speed

54.3

#21 globally

language

69.3

#81 globally

general

42.4

#59 globally

Benchmark Scores

Compare All

Tested on 33 benchmarks · Ranked across 8 categories

Score Distribution (all 312 models)

0255075100

▲ You are here

codingCompare coding →

LiveBench — Coding

Regularly refreshed coding problems that avoid data contamination. New problems added monthly to prevent memorization.

79.7·

SWE-Bench verified

Real-world software engineering tasks from GitHub issues. Models must diagnose bugs and write patches that pass test suites. Human-verified subset of SWE-bench.

78.7·

LiveBench — Agentic Coding

LiveBench coding tasks that require multi-step reasoning and tool use. Tests planning and execution of complex coding workflows.

73.3·

reasoningCompare reasoning →

LiveBench — Reasoning

Regularly refreshed reasoning problems testing logical deduction, spatial reasoning, and analytical thinking.

78.6·

ARC-AGI

Abstraction and Reasoning Corpus. Tests fluid intelligence through novel visual pattern recognition puzzles. Core measure of general intelligence.

77.0·

LiveBench — Data Analysis

Fresh data analysis tasks testing ability to interpret tables, charts, and statistical data.

73.7·

mathCompare math →

LiveBench — Mathematics

Regularly updated math problems that test numerical reasoning, algebra, calculus, and combinatorics.

89.8·

OTIS Mock AIME 2024-2025

Mock AIME (American Invitational Mathematics Exam) problems from OTIS. Tests mathematical competition performance.

86.4·

FrontierMath-Tiers-1-3-v2-Private

59.2·

Quick compare:

vs GPT-5.6 Luna

vs Qwen2 VL 7B Instruct

vs Gemini 3.1 Pro Preview

Excellent (85+) Good (70-85) Average (50-70) Below (<50)

Model Family · z-ai GLM 5.2

GLM 5.2Jun 2026

64.7

$1.00/M in1.0Mctx33 benchmarks

GLM 5.2 (batch)Jun 2026

$0.70/M in(-0.30)1.0Mctx

GLM 5.2 (free)Jun 2026

$0.00/M in(-0.70)33Kctx(-1.0M)

See the full GLM 5.2 family →

Similar Models

Gemini 3.1 Pro Preview

Google DeepMind

65.0$2.00/1M

Links

Info

z-ai Pricing explorer Developers · API

Research

Documentation

Community

Source Code

BenchGecko API

glm-5-2

Specifications

Typetext
Context1.0M tokens (~524 books)
ReleasedJun 2026
LicenseOpen Source
StatusActive
Cost / Message~$0.006

Available On

z-ai$1.00

Frequently Asked Questions

GLM 5.2 is an open-source text AI model by z-ai, released in June 2026. It has an average benchmark score of 64.7. Context window: 1M tokens.

Benchmarks

Chatbot Arena Elo — Coding Chatbot Arena Elo — Overall LiveBench — Mathematics Dtbench GPQA diamond

z-ai · Provider z-ai · Economy All Models Compare Models Pricing Developers · API Context Window · Glossary

Key facts · as of 2026-10-05

GLM 5.2 by z-ai. BenchGecko score 64.7, rank 60 of 312 scored models (normalized average of public benchmark scores).
List price $1.00 input · $4.00 output per 1M tokens (as of 2026-10-05).
Gecko Tests: Who Are You D (Sometimes says it is Google).

How to cite · data as of 2026-10-05

GLM 5.2 · benchmarks, pricing and providers. BenchGecko, data as of 2026-10-05. https://benchgecko.ai/model/glm-5-2

Credit "Source: BenchGecko" with a link. Prices per provider and Gecko Tests are BenchGecko data (CC BY 4.0); benchmark scores keep their original source, listed in the JSON. JSON · llms.txt · MCP

GLM 5.2

Frequently Asked Questions

Related Models

Benchmarks

Related Pages

Key facts · as of 2026-10-05

How to cite · data as of 2026-10-05