How to Use GLM 4.7 Flash in Claude
Use GLM 4.7 Flash for frontend and backend coding, Chinese writing, translation, and long-text review. For code repair, the maker's benchmark settings offer a starting point: temperature 0.7 and top-p 1.0.
Prompts to try with GLM 4.7 Flash in Claude
Copy an example into Claude, or ask in your own words.
Use GLM 4.7 Flash to summarize the discussion above in 300 words. Separate decisions, unresolved questions, and next actions, quoting a supporting passage for each decision.
Use GLM 4.7 Flash to translate the English text above into simplified Chinese for new customers. Keep product names unchanged and explain ambiguous phrases separately.
Use GLM 4.7 Flash to rewrite the Chinese reply above in a calm, direct tone. Preserve every commitment and date, and keep it under 200 Chinese characters.
Use GLM 4.7 Flash with thinking enabled to compare the two proposals above against our stated requirements. Cite the relevant text and list missing evidence before recommending one.
Use GLM 4.7 Flash to role-play a customer confused by a delayed order. Ask one question at a time, use only the supplied policy, and give feedback after five exchanges.
How to set up GLM 4.7 Flash 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 Flash
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-flash", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Claude finds GLM 4.7 Flash on Spiral, shows the price and runs it.
GLM 4.7 Flash providers and prices
Spiral runs GLM 4.7 Flash through OpenRouter. Open a card for today's price, with every fee included.
Up to $0.002134 per run at the default inputs. $0.07865 per 1M input tokens, $0.52 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 | ||
| Logprobs | on or off | ||
| Max tokens | 1 to 117964 | The most output tokens to generate. | |
| Temperature | 0 to 2 | ||
| Top logprobs | 0 to 20 | ||
| Presence penalty | -2 to 2 | ||
| Frequency penalty | -2 to 2 | ||
| Repetition penalty | 0 to 2 |
Every API field of Z.ai: GLM 4.7 Flash is on its tool page: open its card above.
What can you do with GLM 4.7 Flash in Claude?
Work through long text, translations, and Chinese drafts with GLM 4.7 Flash. Supply the source material and request evidence or terminology notes alongside the result.
Repair a reproducible bug
Give GLM 4.7 Flash the failing behavior, relevant code, and expected result. Request a focused patch and verify it with the failing test.
Build a frontend from written requirements
Ask GLM 4.7 Flash to implement a form or search interface with explicit loading, empty, and error states.
Draft backend behavior
Use GLM 4.7 Flash to draft request validation, pagination, and error handling from a written endpoint specification.
Review long documents
Ask GLM 4.7 Flash to extract decisions, unresolved questions, and conflicting requirements from supplied text, with supporting passages.
Write Chinese support replies
Have GLM 4.7 Flash draft a Chinese response using the customer's message and the supplied support policy.
Translate with a terminology guide
Give GLM 4.7 Flash source text, a target language, and approved terminology. Request a translation plus notes on ambiguous phrases.
Practice customer conversations
Use GLM 4.7 Flash to role-play a customer with a defined concern, then review how clearly the response addressed it.
More on GLM 4.7 Flash and Claude
Overview
GLM 4.7 Flash in other agents
Questions about GLM 4.7 Flash in Claude
Can Claude use GLM 4.7 Flash?
Yes. Add Spiral to Claude as an MCP server, and Claude can find and run GLM 4.7 Flash like any built-in tool.
How much does GLM 4.7 Flash cost in Claude?
Up to $0.002134 per run at the default inputs. $0.07865 per 1M input tokens, $0.52 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 Flash in Claude?
No. Spiral runs GLM 4.7 Flash 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 Flash 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 Flash runs?
GLM 4.7 Flash 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 Flash good at?
Z.ai highlights frontend and backend programming, Chinese writing, translation, and long-text processing. Use those strengths for tasks such as code repair, document review, and bilingual drafts.
How much text can GLM 4.7 Flash handle?
The maker specifies a 200K-token context and up to 128K output tokens. Treat these as model specifications, not a promise of that output allowance in every application.
Which settings should I try for code repair?
Start by testing temperature 0.7 and top-p 1.0, the settings Z.ai used for SWE-bench Verified and Terminal Bench. Those evaluations used a 16,384-token generation cap; they do not establish a universal best setting.
When should I use thinking?
Use thinking for debugging, planning, and tasks with several constraints. For a simple wording change, Z.ai recommends disabling thinking to reduce unnecessary reasoning.
Where does GLM 4.7 Flash need extra checking?
Difficult knowledge questions and repository repairs need verification. Z.ai reports 14.4 on Humanity's Last Exam and 59.2 on SWE-bench Verified, so check factual claims and run relevant tests before accepting a patch.
Start using GLM 4.7 Flash 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


