NEWHow AI models draw the world map · Gecko Tests
Beta
Home/Models/GLM 5.2
z-ai logo

GLM 5.2

by z-ai · Released Jun 2026

Open Source1M Context
64.3
avg score
Rank #62
Compare
Better than 81% of all models
Context
1.0M tokens (~524 books)
Input $/1M
$0.62
Output $/1M
$2.42
Type
text
License
Open Source
Benchmarks
34 tested
Data as of
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 34 benchmarks · BenchGecko score 64.3. 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 64.2 (100% as good) at $0.20/1M input · 68% cheaper
Capabilities
coding
61.7
#41 globally
reasoning
60.5
#66 globally
math
66.2
#63 globally
knowledge
49.8
#134 globally
agentic
45.2
#32 globally
speed
54.3
#23 globally
language
69.3
#81 globally
general
42.4
#64 globally
Benchmark Scores
Compare All
Tested on 34 benchmarks · Ranked across 9 categories
Score Distribution (all 319 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:Anthropic logovs Claude Opus 4.7TII logovs Falcon 2 11BOpenAI logovs GPT-5.6 Luna
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Model Family · z-ai GLM 5.2
Jun 2026
64.3
$0.02/M in1.0Mctx34 benchmarks
Jun 2026
$0.70/M in(+0.68)1.0Mctx
Jun 2026
$0.00/M in(-0.70)33Kctx(-1.0M)
See the full GLM 5.2 family →
Similar Models
Anthropic logo
Claude Opus 4.7
Anthropic
64.3$5.00/1M
TII logo
Falcon 2 11B
TII
64.3n/a
OpenAI logo
GPT-5.6 Luna
OpenAI
64.2$0.20/1M
Links
Info
z-aiPricing explorerDevelopers · API
Research
Technical Report
Documentation
API DocsPlayground
Community
@z-ai
Source Code
GitHubHugging Face
BenchGecko API
glm-5-2
Specifications
  • Typetext
  • Context1.0M tokens (~524 books)
  • ReleasedJun 2026
  • LicenseOpen Source
  • StatusActive
  • Cost / Message~$0.004
Available On
z-ai logoz-ai$0.62
Categories
codingreasoningmathknowledgeagenticspeedlanguagegeneral
Learn More
context-windowtransformeropen weightstokens
Share & Export
Tweet
Related Models
Anthropic logoClaude Opus 4.7
64.3
TII logoFalcon 2 11B
64.3
OpenAI logoGPT-5.6 Luna
64.2
Anthropic logoClaude Sonnet 5
64.4
GLM 5.2 is an open-source text AI model by z-ai, released in June 2026. It has an average benchmark score of 64.3. Context window: 1M tokens.

Related Models

Claude Opus 4.7 · AnthropicFalcon 2 11B · TIIGPT-5.6 Luna · OpenAIClaude Sonnet 5 · AnthropicQwen2.5 Coder 14B Instruct · Alibaba

Benchmarks

Chatbot Arena Elo — CodingChatbot Arena Elo — OverallLiveBench — MathematicsDtbenchGPQA diamond

Related Pages

z-ai · Providerz-ai · EconomyAll ModelsCompare ModelsPricingDevelopers · APIContext Window · Glossary

Key facts · as of 2026-10-16

  • GLM 5.2 by z-ai. BenchGecko score 64.3, rank 60 of 319 scored models (normalized average of public benchmark scores).
  • List price $0.62 input · $2.42 output per 1M tokens (as of 2026-10-16).
  • Sold by 34 providers (as of 2026-10-16): Relace $0.0351 in / $6.00 out · Wafer (fp4) $0.0370 in / $7.00 out · Morph (fp8) $0.0790 in / $6.00 out · Cloudflare $0.19 in / $8.00 out · Wafer (fp4) $0.19 in / $8.00 out · and 29 more. Every provider
  • Gecko Tests: Who Are You D (Sometimes says it is Google) · Tokenizer Tax D (70% more tokens outside English).

How to cite · data as of 2026-10-16

GLM 5.2 · benchmarks, pricing and providers. BenchGecko, data as of 2026-10-16. 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