GLM 5
GLM-5 is Zhipu's (Z.ai) flagship 744B sparse MoE model activating just 40B parameters per token, coding at parity with Claude Opus 4.5. Access it through Velokey's unified endpoint with below-list pricing.
Use cases
Full-stack code generation · Autonomous agents & tool use
Input
Text
Output
Text
Billing
Per token
Start a conversation
Type a message below to begin
Sign up to get $0.5 in free credits — about 20 free images
Pricing details
Transparent usage-based pricing with no hidden fees.
Detailed Pricing
Usage-based pricing breakdown
| Spec | Price | Official | Savings |
|---|---|---|---|
| Input/1M tokens | 0.8 Credits≈$0.8 | 1 Credits | -20% |
| Output/1M tokens | 2.56 Credits≈$2.56 | 3.2 Credits | -20% |
| Cache Read/1M tokens | 0.16 Credits≈$0.16 | 0.2 Credits | -20% |
Billing Rules
- Output price varies by model, resolution, quality, duration, or token usage.
- Uploaded reference files may be charged separately when supported.
- Add-on capabilities such as web search or image search may be billed per request.
- Token-based models are billed by input and output tokens.
- If the primary route is unavailable, Velokey may automatically switch to a stable fallback route when possible.
- Usage is pay-as-you-go with transparent credit deduction.
* Final cost depends on the generated result.
Related models
Explore other models in the same family.
ZhipuSave 20%GLM 5.2
GLM-5.2 is Z.ai's flagship model with a 1M-token context, strong project-level coding, stable multi-step execution, and adjustable thinking for long-horizon engineering.
- ZhipuSave 20%
GLM 5.1
GLM-5.1 is Zhipu's flagship AI model, excelling in programming, complex reasoning, and long-chain tasks.
OpenAISave 20%GPT 5.4 Mini
OpenAI's fast, cost-efficient GPT-5.4 Mini for high-volume production, with vision, long context, and strong tool use at low latency.
AlibabaSave 20%Qwen3.6 Plus
Alibaba's Qwen3.6 Plus, a balanced flagship-tier model with strong reasoning, long context, multimodal input, and agentic tool use.
Affordable GLM-5 API
Access Zhipu's GLM-5 on Velokey — Opus-4.5-level coding, SOTA open-weight agentic tool use, and an efficient 744B/40B sparse architecture, with a unified endpoint and below-list pricing for production apps.
Meet GLM-5 and its API access
GLM-5 is Zhipu's (Z.ai) flagship model, built on a 744B-parameter sparse mixture-of-experts architecture that activates only 40B parameters per token and integrates DeepSeek Sparse Attention for efficiency at scale. Pre-trained on 28.5T tokens and refined with the async "Slime" reinforcement-learning framework, it codes at parity with Claude Opus 4.5, ranks first among open-weight models on agentic benchmarks, and supports multiple thinking modes. Velokey provides an affordable, easy-to-integrate GLM-5 API: a unified chat endpoint, a clear request flow, and below-list pricing for production systems, agent pipelines, and automation tools.
Opus-4.5-level coding
Generates runnable code from natural language across frontend, backend, and data processing, reaching performance parity with Claude Opus 4.5 on software engineering — 77.8 on SWE-bench Verified and 56.2 on Terminal Bench 2.0.
View Documentation →
SOTA open-weight agentic tool use
Autonomous decision-making and tool invocation turn a single sentence into a complete deliverable through multi-step execution, ranking first among open-weight models on BrowseComp, MCP-Atlas, and τ²-Bench.
View Documentation →
Efficient 744B/40B sparse architecture
744B total parameters activate only 40B per token, combined with DeepSeek Sparse Attention (DSA) and pre-training on 28.5T tokens for frontier quality at lower serving cost.
View Documentation →
Thinking modes and rich tooling
Multiple thinking modes trained with the async "Slime" RL framework, plus function calling, structured JSON output, context caching, and streaming for reliable production integration.
View Documentation →
How to deploy the GLM-5 API on Velokey
Get started with just a few steps.
Sign up and get an API Key
Sign in to the Velokey console to generate an API Key that authenticates every request sent to GLM-5.
Configure request parameters
Set model to glm-5 and submit messages. Select a thinking mode for deeper reasoning, and enable function calling, structured JSON output, and context caching.
Integrate and start calling
Send requests to Velokey's unified /v1/chat/completions endpoint to power assistants, agents, and automation, with usage managed in one console.
Real-world use cases for GLM-5
From full-stack engineering to autonomous agents, covering many high-autonomy scenarios.
Full-stack code generation
With Opus-4.5-level coding, fit for generating runnable frontend, backend, and data-processing code from natural language across the whole stack.
Autonomous agents & tool use
Ranks first among open-weight models on agentic benchmarks — fit for browsing, MCP tool orchestration, and multi-step task automation.
Complex reasoning & research
Multiple thinking modes trained with async RL — fit for deep reasoning, analysis, and long-horizon research workflows.
Data processing pipelines
Generates runnable data-processing code from natural language — fit for ETL, transformation, and analytics automation.
Structured output & integration
Function calling and structured JSON output — fit for wiring the model into backend services and typed data flows.
Long-context analysis
A 200K-token context and up to 128K output — fit for analyzing large documents, codebases, and long conversations.
Why choose Velokey for GLM-5
Discount on list price
Call GLM-5 at a lower price on Velokey — pay-as-you-go with no forced subscription, reducing upfront cost.
Unified API interface
One API Key accesses GLM-5 and other models — no juggling multiple accounts and keys, simplifying integration.
Complete docs and integration guidance
Endpoint references, parameter details, and examples help you adopt thinking modes, function calling, and structured output smoothly.
Stable and highly available
Automatic failover and stable fallback routes maintain availability when the primary route is down, keeping production reliable.
API Reference
Complete API call examples and parameter descriptions
Endpoint
https://api.velokey.ai/v1/chat/completionsAuthentication
Bearer YOUR_API_KEYRequest Example
curl https://api.velokey.ai/v1/chat/completions \
-X POST \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "glm-5",
"messages": [{"role": "user", "content": "Hello!"}],
"temperature": 0.7
}'Response Example
{
"id": "req_abc123",
"model": "glm-5",
"created": 1234567890,
"data": { ... }
}GLM-5 API FAQ
What is GLM-5's architecture?
GLM-5 uses a 744B-parameter sparse mixture-of-experts architecture that activates only 40B parameters per token, integrates DeepSeek Sparse Attention for efficiency, and was pre-trained on 28.5T tokens, then refined with the async "Slime" RL framework.
How good is GLM-5 at coding?
It generates runnable code from natural language across frontend, backend, and data processing, reaching performance parity with Claude Opus 4.5 on software engineering, scoring 77.8 on SWE-bench Verified and 56.2 on Terminal Bench 2.0.
How strong are its agentic capabilities?
GLM-5 performs autonomous decision-making and tool invocation, turning a single sentence into a complete deliverable through multi-step execution. It ranks first among open-weight models on BrowseComp, MCP-Atlas, and τ²-Bench.
What context and output limits does it support?
GLM-5 supports a 200K-token context window and up to 128K tokens of output. It handles text-to-text tasks with multiple thinking modes, function calling, structured JSON output, context caching, and streaming.
How is it billed on Velokey?
Velokey routes GLM-5 through a unified endpoint with one API key at below-list pricing, billed by input and output tokens. This page's pricing area shows the current Velokey price in the dashboard.
Start building with GLM 5 today
Use one API key to access low-cost, stable AI models through Velokey.