How to Use GLM-4.7 in Claude
GLM-4.7 suits code repairs, web UI work, and tasks that require several tool calls. Compared with GLM-4.6, it improves multilingual coding and terminal tasks and adds control over thinking on each turn.
Prompts to try with GLM-4.7 in Claude
Copy an example into Claude, or ask in your own words.
Use GLM-4.7 to compare the proposals in this conversation. List agreements, contradictions, and unanswered questions, citing the relevant passage for each.
Use GLM-4.7 with thinking enabled to plan a three-stage product launch. Identify dependencies, decision points, and assumptions that need evidence.
Use GLM-4.7 to extract action items from the meeting transcript in this conversation. Return JSON containing the action, owner, due date, and supporting quotation; use null for missing details.
Use GLM-4.7 to draft HTML for a five-slide presentation explaining the proposal above. Give each slide one main point and a consistent layout.
Use GLM-4.7 to write a branching dialogue between a cautious engineer and an impatient expedition leader. Keep their motives consistent and offer three meaningful choices.
How to set up GLM-4.7 in Claude
Pick the way that suits you. Each takes about two minutes.
Use the Claude connector
- Add Spiral to Claude
Click Add Spiral in Claude above. Claude opens Add custom connector with Spiral and its URL filled in; check them and click Continue. Without the link: in Claude open Customize › Connectors › Add › Add custom connector and paste this URL:
https://api.spiral.ai/mcp
- Keep the detected sign-in and click Add
Claude finds Spiral's sign-in by itself: leave Sign in now and Use Claude's published identity selected.
- Click Connect
The connector page opens. Connect takes you to Spiral in your browser.
Use the Spiral skill
- Send Claude 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-4.7
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-4.7", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Claude finds GLM-4.7 on Spiral, shows the price and runs it.
GLM-4.7 providers and prices
Spiral runs GLM-4.7 through OpenRouter. Open a card for today's price, with every fee included.
Up to $0.009337 per run at the default inputs. $0.52 per 1M input tokens, $2.275 per 1M output tokens.
Settings you can ask for
Ask Claude for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Seed | a whole number | ||
| Min p | 0 to 1 | ||
| Top k | a whole number from 0 | ||
| Top p | 0 to 1 | ||
| Max tokens | 1 to 131072 | 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 4.7 is on its tool page: open its card above.
What can you do with GLM-4.7 in Claude?
GLM-4.7 can turn a detailed question into a research brief, structured record, or creative draft. Supply the source text and describe the result you need.
Repair a reproducible bug
Give GLM-4.7 the failure, relevant code, and expected behavior. Request a focused patch and a test that reproduces the problem.
Build an interface from a written brief
Ask GLM-4.7 for responsive interface code with explicit layout, content, and interaction requirements.
Diagnose terminal failures
Have GLM-4.7 investigate build or test failures, use command results to revise its diagnosis, and report what remains unresolved.
Draft presentation code
Turn an outline into HTML slides with GLM-4.7. Specify dimensions, slide count, and a consistent heading hierarchy.
Consolidate research
Ask GLM-4.7 to compare supplied source text, identify disagreements, and produce a brief with evidence attached to each conclusion.
Extract records from text
Give GLM-4.7 an explicit JSON structure and request records extracted from messages or descriptions, with missing values marked consistently.
Develop interactive fiction
Use GLM-4.7 to draft dialogue and branching scenes from a character brief, then revise against stated continuity rules.
More on GLM-4.7 and Claude
Overview
GLM-4.7 in other agents
Questions about GLM-4.7 in Claude
Can Claude use GLM-4.7?
Yes. Add Spiral to Claude as an MCP server, and Claude can find and run GLM-4.7 like any built-in tool.
How much does GLM-4.7 cost in Claude?
Up to $0.009337 per run at the default inputs. $0.52 per 1M input tokens, $2.275 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Claude plan is separate.
Do I need an OpenRouter account to use GLM-4.7 in Claude?
No. Spiral runs GLM-4.7 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-4.7 in Claude?
About two minutes: add Spiral to Claude, then keep the detected sign-in and click Add, then click Connect.
Does Claude wait while GLM-4.7 runs?
GLM-4.7 runs in the background. Claude starts the run, checks on it and gives you the text, json when it's ready.
What is GLM-4.7 good at?
GLM-4.7 is a useful choice for coding tasks that involve several steps, generated interfaces, and research synthesis. Give it a concrete deliverable and explicit acceptance criteria.
How much text can GLM-4.7 handle?
Z.ai documents a 200K-token context window and up to 128K output tokens. Treat those as ceilings and request only the output your task needs.
Can GLM-4.7 read screenshots or generate images?
Its documented input and output are text. For visual work, provide a written design specification and ask for interface or slide code.
When should I ask GLM-4.7 to use thinking?
Use thinking for debugging, planning, and decisions that depend on tool results. GLM-4.7 supports switching thinking on or off for individual turns.
Which sampling settings should I try for coding?
A concrete starting point is Z.ai’s coding evaluation configuration: temperature 0.7, top-p 1.0, and a 16,384-token generation limit. These are published test settings, so compare them against your own tasks.
Start using GLM-4.7 in Claude
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into Claude:
set up https://spiral.ai/SKILL.md


