How to Use GLM 4.5 Air in Muse Code
GLM 4.5 Air handles coding, tool use, and reasoning tasks. It uses a smaller design than GLM 4.5, with 106 billion total parameters and 12 billion active parameters.
Prompts to try with GLM 4.5 Air in Muse Code
Copy an example into Muse Code, or ask in your own words.
Use GLM 4.5 Air with thinking enabled to investigate the failing tests in this project. Reproduce the failure, trace its cause, make the smallest useful fix, and run the relevant checks.
Use GLM 4.5 Air to build a keyboard-accessible filter panel using this project's existing components. Include empty and error states, then check the result in the browser.
Use GLM 4.5 Air with thinking enabled to trace login from the route handler to session storage. Cite file paths and explain where an expired session is rejected in under 600 words.
Use GLM 4.5 Air to refactor the duplicated date validation in this project. Preserve current behavior and verify valid dates, leap days, and invalid input.
Use GLM 4.5 Air to review the current diff for incorrect error handling. Report only findings with a concrete failure scenario and a file location; keep the review under 800 words.
How to set up GLM 4.5 Air in Muse Code
Pick the way that suits you. Each takes about two minutes.
Use Spiral's MCP server
- Install Muse Code
curl -fsSL https://dev.meta.ai/install.sh | bash
- Ask Muse to add the server
Start muse and ask:
Add the Spiral MCP server at https://api.spiral.ai/mcp
- Sign in
In your terminal, run:
muse mcp login spiral
Use the Spiral skill
- Send Muse Code this message
Your agent reads Spiral's skill and sets itself up.
set up https://spiral.ai/SKILL.md
- Sign in or paste a key
Approve Spiral in the browser, or give the agent a key from the console.
Use the Spiral CLI
- Install and sign in
Installs the Spiral CLI, signs you in and adds its skill to your coding agents.
npx -y spiral-cli@latest init
Call the Spiral API
- Create an API key
In the console, under API keys.
- Run GLM 4.5 Air
One request; the result comes back by polling or a webhook.
curl https://api.spiral.ai/v1/runs \ -H "Authorization: Bearer $SPIRAL_API_KEY" \ -H "Content-Type: application/json" \ -H "Idempotency-Key: $(uuidgen)" \ -d '{ "provider": "openrouter", "model": "z-ai/glm-4.5-air", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Muse Code finds GLM 4.5 Air on Spiral, shows the price and runs it.
GLM 4.5 Air providers and prices
Spiral runs GLM 4.5 Air through OpenRouter. Open a card for today's price, with every fee included.
Up to $0.004534 per run at the default inputs. $0.169 per 1M input tokens, $1.105 per 1M output tokens.
Settings you can ask for
Ask Muse Code for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Seed | a whole number | ||
| Top k | a whole number from 0 | ||
| Top p | 0 to 1 | ||
| Max tokens | 1 to 98304 | The most output tokens to generate. | |
| Temperature | 0 to 2 | ||
| Presence penalty | -2 to 2 | ||
| Frequency penalty | -2 to 2 | ||
| Repetition penalty | 0 to 2 |
Every API field of Z.ai: GLM 4.5 Air is on its tool page: open its card above.
What can you do with GLM 4.5 Air in Muse Code?
Turn a bug report or written feature brief into code you can check in your project. GLM 4.5 Air can work through coding tasks and use development tools to inspect the result.
Fix a reproducible bug
Give GLM 4.5 Air the failure, relevant code, and expected behavior. Ask for a focused patch and a regression check.
Build an interface from a written brief
Ask GLM 4.5 Air to turn requirements into working front-end code, including loading, empty, and error states.
Explain code before changing it
Have GLM 4.5 Air trace a request through the relevant functions and identify where a proposed change belongs.
Research a technical decision
Give GLM 4.5 Air a specific question and access to search tools. Request a comparison supported by primary sources.
Check a plan against constraints
Ask GLM 4.5 Air to evaluate dependencies, capacity, and deadlines, then identify which assumptions could make the plan fail.
Extract consistent records
Give GLM 4.5 Air a target JSON structure and source text. Require evidence for each extracted value and null for missing information.
More on GLM 4.5 Air and Muse Code
Overview
GLM 4.5 Air in other agents
More Muse Code guides
Questions about GLM 4.5 Air in Muse Code
Can Muse Code use GLM 4.5 Air?
Yes. Add Spiral to Muse Code as an MCP server, and Muse Code can find and run GLM 4.5 Air like any built-in tool.
How much does GLM 4.5 Air cost in Muse Code?
Up to $0.005867 per run at the default inputs. $0.26 per 1M input tokens, $1.43 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Muse Code plan is separate.
Do I need an OpenRouter account to use GLM 4.5 Air in Muse Code?
No. Spiral runs GLM 4.5 Air on its own OpenRouter account. You sign in to Spiral once and never paste a key.
How long does it take to set up GLM 4.5 Air in Muse Code?
About two minutes: install Muse Code, then ask Muse to add the server, then sign in.
Does Muse Code wait while GLM 4.5 Air runs?
GLM 4.5 Air runs in the background. Muse Code starts the run, checks on it and gives you the text, json when it's ready.
What is GLM 4.5 Air good at?
Its documented strengths include software engineering, front-end development, reasoning, and tool use. Try it for debugging, building a working interface from a written brief, or researching a question through an agent's search tools.
How much text can GLM 4.5 Air handle?
Z.ai documents a 128K context window and up to 96K output tokens. Treat these as ceilings; give each task a shorter target, such as a 500-word brief or one patch.
When should I use thinking mode?
Use thinking mode for difficult debugging, reasoning, and tasks involving several tools. Use non-thinking mode for straightforward requests such as labeling a short message or rewriting a sentence.
Can GLM 4.5 Air analyze images or generate video?
Z.ai documents text input and text output. For a visual task, provide a written description or extracted text and ask for code, analysis, or a script.
Which languages should I start with?
The model card lists English and Chinese. Start evaluations in those languages and check terminology against examples from your own work.
Start using GLM 4.5 Air in Muse Code
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into Muse Code:
set up https://spiral.ai/SKILL.md


