How to Use GLM-4.6 in Claude
GLM-4.6 suits code changes, front-end development, and research that uses tools. Its 200K-token context supports long text tasks, while Z.ai reports that its coding ability trails Claude Sonnet 4.5.
Prompts to try with GLM-4.6 in Claude
Copy an example into Claude, or ask in your own words.
Use GLM-4.6 to compare the two proposals pasted above. Return a table of agreements, conflicts, and missing decisions, with a supporting passage for each conflict.
Use GLM-4.6 to rewrite the draft above for first-time customers in 250 words or fewer. Preserve its factual claims, remove repetition, and explain any unclear terms.
Use GLM-4.6 to translate the product description above into Japanese for an online store. Preserve product names and measurements, follow the supplied glossary, and flag ambiguous phrases.
Use GLM-4.6 to investigate the claims in the announcement above using available search tools. Link supporting sources and mark claims you cannot verify.
Use GLM-4.6 to examine the code and error text pasted above. Explain the likely cause, suggest a minimal correction, and give one example that checks the fix.
How to set up GLM-4.6 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-4.6
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-4.6", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Claude finds GLM-4.6 on Spiral, shows the price and runs it.
GLM-4.6 providers and prices
Spiral runs GLM-4.6 through OpenRouter. Open a card for today's price, with every fee included.
Up to $0.010673 per run at the default inputs. $0.65 per 1M input tokens, $2.60 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 4.6 is on its tool page: open its card above.
What can you do with GLM-4.6 in Claude?
Use GLM-4.6 to compare long passages, revise drafts, and translate product copy. State the audience, desired format, and evidence the answer should use.
Repair a failing feature
Give GLM-4.6 the relevant code, error output, and expected behavior. Ask for a focused patch and a regression check.
Build a front-end from a written brief
Ask GLM-4.6 to turn layout requirements into interface code, including narrow-screen behavior, empty states, and error messages.
Compare long specifications
Supply labeled specification text and ask GLM-4.6 for conflicting requirements, missing decisions, and supporting section references.
Prepare a research brief
Have GLM-4.6 use available search tools to investigate a narrow question, compare sources, and separate supported conclusions from open questions.
Edit customer-facing copy
Give GLM-4.6 a draft, audience, and tone guide. Request a clearer version that preserves the supplied claims and removes repetition.
Translate product descriptions
Ask GLM-4.6 for French, Russian, Japanese, or Korean product copy using a supplied glossary, with product names and measurements preserved.
Draft a recurring change digest
Have GLM-4.6 compare retrieved release-note text with the previous digest and explain changes, affected users, and unresolved questions.
More on GLM-4.6 and Claude
Overview
GLM-4.6 in other agents
Questions about GLM-4.6 in Claude
Can Claude use GLM-4.6?
Yes. Add Spiral to Claude as an MCP server, and Claude can find and run GLM-4.6 like any built-in tool.
How much does GLM-4.6 cost in Claude?
Up to $0.010673 per run at the default inputs. $0.65 per 1M input tokens, $2.60 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-4.6 in Claude?
No. Spiral runs GLM-4.6 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-4.6 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-4.6 runs?
GLM-4.6 runs in the background. Claude starts the run, checks on it and gives you the text, json when it's ready.
What is GLM-4.6 useful for?
Use GLM-4.6 for code changes, front-end development, and research that combines reasoning with tools. Give it a concrete deliverable and acceptance criteria.
How much text can GLM-4.6 handle?
Z.ai documents a 200K-token context window and a maximum output of 128K tokens. Treat those as ceilings and request only the output your task needs.
Can GLM-4.6 read screenshots or generate images?
GLM-4.6 has text input and text output. For a visual debugging task, supply the code, error text, and a written description of the problem.
Which settings should I try for coding?
Z.ai recommends top_p of 0.95 and top_k of 40 for code evaluations such as LiveCodeBench. Use those as a starting point and compare results on your own tasks.
Which translation languages have documented improvements?
Z.ai specifically names French, Russian, Japanese, and Korean. Include the target audience, tone, and a glossary when requesting a translation.
Start using GLM-4.6 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


