Home/Models/GLM 4.6
z-ai logo

GLM 4.6

by z-ai · Released Sep 2025

Open Source
54.4
avg score
Rank #91
Compare
Better than 61% of all models
Context
205K tokens (~102 books)
Input $/1M
$0.39
Output $/1M
$1.90
Type
text
License
Open Source
Benchmarks
20 tested
Data updated today
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 with 50.8% average. Top scores: Chatbot Arena Elo — Overall (1425.8%), Chatbot Arena Elo — Coding (1353.7%), OpenCompass — AIME2025 (90.3%).

Looking for similar performance at lower cost?
Phi 4 scores 54.2 (100% as good) at $0.07/1M input · 83% cheaper
Capabilities
coding
52.2
#61 globally
reasoning
57.0
#38 globally
math
44.3
#95 globally
knowledge
59.5
#51 globally
agentic
3.0
#37 globally
language
58.0
#91 globally
Benchmark Scores
Compare All
Tested on 20 benchmarks · Ranked across 7 categories
Score Distribution (all 233 models)
0255075100
▲ You are here
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
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
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 Problems are unpublished and cannot be memorized.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
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Links
Documentation
Community
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$0.39
Share & Export
Tweet
GLM 4.6 is an open-source text AI model by z-ai, released in September 2025. It has an average benchmark score of 54.4. Context window: 205K tokens.