Z.ai GLM 5
Z.aiBalanced
ThinkingTool UseStructured Output
About this model
GLM-5 is Z.ai’s flagship open-source foundation model engineered for complex systems design and long-horizon agent workflows. Built for expert developers, it delivers production-grade performance on large-scale programming tasks, rivaling leading...
Performance Tier
Balanced
Z.ai GLM 5 is a balanced model from Z.ai : strong performance at a reasonable price.
Strong cost-performance ratio. Reliable for most professional use cases without premium pricing.
Pricing
This model is included in Elosia plansAffordable
Low cost. Suitable for sustained use and high-volume interactions.
| Type | per 1M tokens |
|---|---|
| Input (prompt) | $0.950 |
| Output (completion) | $2.55 |
| Cache read | $0.200 |
Capabilities
Context Length205K
Max Output Tokens131K
TokenizerOther
Inputtext
Outputtext
Release DateFebruary 11, 2026
Benchmarks
General Intelligence
MMLU
Not reported
GPQA Diamond
86%
Mathematics
MATH-500
Not reported
Programming
HumanEval
Not reported
SWE-bench Verified
77.8%
Agentic
Terminal-Bench 2.0
56.2%
Where does Z.ai GLM 5 stand?
Compare its performance index against every other model.
Recommended Use Cases
CodingAnalysisResearchGeneral Chat
Strengths
- 744B MoE model (40B active) — competitive with Claude Opus and GPT-5
- 200K context window with up to 128K output tokens
- Strong agentic capabilities (BrowseComp #1, MCP-Atlas #1 among open-weight)
- Open-weight MIT license — self-hostable
- Extremely cost-efficient vs comparable frontier models
Limitations
- Full 744B MoE requires heavy compute to self-host despite open weights
- Smaller ecosystem of integrations outside Chinese market
Frequently asked questions
Resources
This model may use your data for training
