GLM 4.6 FP8
GLM 4.6 is a chat model built by Z.ai. It is genuinely best at advanced reasoning, coding, and agentic tasks, with notable improvements in generating polished front-end pages.
Input
Output
Context
203K
Max Output
128K
Parameters
358.5B
Technical Specifications
Model TypeChat
Context Window202,752 tokens
Max Output Tokens128,000 tokens
Parameters358.5B
Release DateSep 29, 2025
Training CutoffApr 1, 2025
Licensemit
Open SourceYes
Input Modalities
Text
Output Modalities
Text
Capabilities
Benchmarks
Artificial Analysis
13.3%
HLE
54.3%
AA-LCR
78.0%
GPQA
70.5%
Tau2-Bench
86.0%
AIME 2025
43.4%
IFBench
38.4%
SciCode
82.9%
MMLU Pro
86.0
Math
10.5%
Tau2 Banking
45.8
Coding
69.5%
LiveCodeBench
28.7
Intelligence
25.0%
Terminal-Bench Hard
49.4%
Terminal-Bench 2.1
Estimated GPU Requirements for GLM 4.6 FP8
Estimates based on INT8 quantization. Actual requirements vary by framework and configuration.
More Models by Z.ai
Related Tools
Data sourced from official provider APIs and documentation
Last updated: Aug 6, 2026
Your AI stack shouldn't stand still.
Every month new models become cheaper, faster, and more capable. Inferbase ensures your application automatically benefits without changing a single API call.