NEWHow AI models draw the world map · Gecko Tests
Beta
Home/Models/GLM 4.6
z-ai logo

GLM 4.6

by z-ai · Released Sep 2025

Open Source
53.6
avg score
Rank #123
Compare
Better than 61% of all models
Context
205K tokens (~102 books)
Input $/1M
$0.43
Output $/1M
$1.75
Type
text
License
Open Source
Benchmarks
20 tested
Data as of
About

Compared with GLM-4.5, this generation brings several key improvements: Longer context window: The context window has been expanded from 128K to 200K tokens, enabling the model to handle more complex...

Tested on 20 benchmarks · BenchGecko score 53.6. Top scores: Chatbot Arena Elo — Overall (1424.1%), Chatbot Arena Elo — Coding (1339.4%), OpenCompass — AIME2025 (90.3%).

Looking for similar performance at lower cost?
Gemma 4 31B scores 53.3 (99% as good) at $0.09/1M input · 79% cheaper
Capabilities
coding
52.2
#83 globally
reasoning
57.0
#67 globally
math
44.3
#139 globally
knowledge
59.5
#60 globally
agentic
3.0
#70 globally
language
58.0
#109 globally
Benchmark Scores
Compare All
Tested on 20 benchmarks · Ranked across 7 categories
Score Distribution (all 312 models)
0255075100
▲ You are here
codingCompare coding →
OpenCompass — LiveCodeBenchV6

OpenCompass Live Code Bench v6. Fresh competitive programming problems to evaluate code generation without memorization.

78.2·
LiveBench — Coding

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

71.0·
LiveBench — Agentic Coding

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

35.0·
reasoningCompare reasoning →
LiveBench — Reasoning

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

62.1·
LiveBench — Data Analysis

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

52.0·
mathCompare math →
OpenCompass — AIME2025

OpenCompass evaluation on AIME 2025 problems. Tests mathematical reasoning on fresh competition problems.

90.3·
LiveBench — Mathematics

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

81.1·
FrontierMath-2025-02-28-Private

Original research-level math problems created by professional mathematicians. Problems are unpublished and cannot be memorized.

3.8·
Quick compare:Alibaba Qwen logovs Qwen3.5-35B-A3Bmeituan logovs LongCat Flash ChatGoogle DeepMind logovs Gemma 4 31B
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Recently Happened
GLM 4.6 pricing dropped 22%
Sep 9, 2026
GLM 4.6 pricing increased 28%
Sep 3, 2026
GLM 4.6 pricing dropped 14%
Aug 27, 2026
GLM 4.6 pricing dropped 9%
Aug 17, 2026
GLM 4.6 pricing increased 10%
Aug 15, 2026
Similar Models
Alibaba Qwen logo
Qwen3.5-35B-A3B
Alibaba Qwen
53.5$0.15/1M
meituan logo
LongCat Flash Chat
meituan
53.4$0.20/1M
Google DeepMind logo
Gemma 4 31B
Google DeepMind
53.3$0.09/1M
Links
Info
z-aiPricing explorerDevelopers · API
Research
Technical Report
Documentation
API DocsPlayground
Community
@z-ai
Source Code
GitHubHugging Face
BenchGecko API
glm-4-6
Specifications
  • Typetext
  • Context205K tokens (~102 books)
  • ReleasedSep 2025
  • LicenseOpen Source
  • StatusActive
  • Cost / Message~$0.003
Available On
z-ai logoz-ai$0.43
Categories
codingreasoningmathknowledgeagenticlanguage
Learn More
context-windowtransformeropen weightstokens
Share & Export
Tweet
Related Models
Alibaba Qwen logoQwen3.5-35B-A3B
53.5
meituan logoLongCat Flash Chat
53.4
Google DeepMind logoGemma 4 31B
53.3
DeepSeek logoDeepSeek R1 Distill Llama 8B
53.1
GLM 4.6 is an open-source text AI model by z-ai, released in September 2025. It has an average benchmark score of 53.6. Context window: 205K tokens.

Related Models

Qwen3.5-35B-A3B · Alibaba QwenLongCat Flash Chat · meituanGemma 4 31B · Google DeepMindDeepSeek R1 Distill Llama 8B · DeepSeekGrok 3 Mini Beta · xAI

Benchmarks

Chatbot Arena Elo — OverallChatbot Arena Elo — CodingOpenCompass — AIME2025OpenCompass — IFEvalOpenCompass — MMLU-Pro

Related Pages

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

Key facts · as of 2026-10-05

  • GLM 4.6 by z-ai. BenchGecko score 53.6, rank 123 of 312 scored models (normalized average of public benchmark scores).
  • List price $0.43 input · $1.75 output per 1M tokens (as of 2026-10-05).
  • Sold by 4 providers (as of 2026-10-05): Venice (fp4) $0.43 in / $1.75 out · DeepInfra (fp4) $0.50 in / $2.00 out · Novita (bf16) $0.55 in / $2.20 out · Z.AI (fp4) $0.60 in / $2.20 out. Every provider

How to cite · data as of 2026-10-05

GLM 4.6 · benchmarks, pricing and providers. BenchGecko, data as of 2026-10-05. https://benchgecko.ai/model/glm-4-6

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