How to Use GLM 5.3 FlashX in Devin
GLM 5.3 FlashX suits coding, visual checks, and document analysis in tool-using workflows. Compared with GLM 5.3 Flash, it emphasizes faster inference while retaining multimodal capabilities.
Prompts to try with GLM 5.3 FlashX in Devin
Copy an example into Devin, or ask in your own words.
Use GLM 5.3 FlashX to implement the settings page described in this project's product brief. Follow the existing components, test the main interactions, and inspect the rendered page before reporting remaining issues.
Use GLM 5.3 FlashX to compare the reference screenshots in this repository with the running app. Fix the three largest layout differences and capture the result for review.
Use GLM 5.3 FlashX with maximum reasoning effort to trace the failing checkout flow. Read the relevant code, reproduce the failure, make a focused fix, and verify the full interaction.
Use GLM 5.3 FlashX to build a document extraction workflow in this project. Define the expected JSON structure, preserve source excerpts, and validate required fields before accepting a result.
Use GLM 5.3 FlashX to create a presentation from the research notes in this repository. Use the available document tools, render each slide, and fix overflow or overlapping elements.
How to set up GLM 5.3 FlashX in Devin
Pick the way that suits you. Each takes about two minutes.
Use Spiral's MCP server
- Add a custom MCP
Open Customize › MCPs. Click + and choose New MCP. Fill in:
- Sign in to Spiral and approve
In Customize › MCPs, open Spiral and choose Connect. Spiral opens: sign in, pick the organization and click Authorize. No API key.
- Check it
On Spiral's configuration page, click Test tools. Devin lists Spiral's tools.
Use the Spiral skill
- Send Devin 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 5.3 FlashX
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-5.3-flashx", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Devin finds GLM 5.3 FlashX on Spiral, shows the price and runs it.
GLM 5.3 FlashX providers and prices
Spiral runs GLM 5.3 FlashX through OpenRouter. Open a card for today's price, with every fee included.
Up to $0.006673 per run at the default inputs. $0.481 per 1M input tokens, $1.625 per 1M output tokens.
Settings you can ask for
Ask Devin for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| 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 | ||
| Reasoning effort | low, high, or max | max | How much the model reasons. |
Every API field of Z.ai: GLM 5.3 FlashX is on its tool page: open its card above.
What can you do with GLM 5.3 FlashX in Devin?
GLM 5.3 FlashX can take a coding task through implementation and visual review. Give the agent project context and clear acceptance checks, then ask it to inspect the running result.
Implement a product interface
Turn a written product brief into frontend code, then use browser tools to inspect the result and revise visible problems.
Review a rendered interface
Compare interface screenshots with the intended design and produce a prioritized list of layout and interaction fixes.
Build a management presentation
Organize supplied business material into a slide narrative, then use document tools to create an editable presentation.
Check report layout
Inspect rendered report pages for cropped content, overlapping elements, and inconsistent alignment before revising the document.
Synthesize source material
Combine public information and supplied evidence into a research brief that separates reported facts, assumptions, and conclusions.
Extract document data
Convert document text into a specified JSON structure, with explicit handling for missing values and source excerpts for verification.
More on GLM 5.3 FlashX and Devin
Overview
GLM 5.3 FlashX in other agents
Questions about GLM 5.3 FlashX in Devin
Can Devin use GLM 5.3 FlashX?
Yes. Add Spiral to Devin as an MCP server, and Devin can find and run GLM 5.3 FlashX like any built-in tool.
How much does GLM 5.3 FlashX cost in Devin?
Up to $0.006673 per run at the default inputs. $0.481 per 1M input tokens, $1.625 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Devin plan is separate.
Do I need an OpenRouter account to use GLM 5.3 FlashX in Devin?
No. Spiral runs GLM 5.3 FlashX 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 5.3 FlashX in Devin?
About two minutes: add a custom MCP, then sign in to Spiral and approve, then check it.
Does Devin wait while GLM 5.3 FlashX runs?
GLM 5.3 FlashX runs in the background. Devin starts the run, checks on it and gives you the text, json when it's ready.
What is GLM 5.3 FlashX good at?
Use it for coding, visual review, and document workflows that involve several tool-assisted steps. Tasks with a clear result to inspect, such as a rendered interface or report, fit its visual feedback approach.
How does GLM 5.3 FlashX differ from GLM 5.3 Flash?
GLM 5.3 FlashX emphasizes faster inference for the multimodal coding model. Treat speed as the main documented distinction when choosing between them.
How much can GLM 5.3 FlashX read and write?
Z.ai lists a 1M-token context window and up to 128K output tokens. Its native output is text; workflows producing presentations or spreadsheets require tools to create those files.
Can GLM 5.3 FlashX skip reasoning?
Z.ai documents forced thinking for GLM 5.3 Flash, including the FlashX offering described on its model page. For brief tasks, request a concise final answer instead of assuming reasoning can be switched off.
Which settings should I start with?
Start with Z.ai's recommended temperature of 1, top-p of 0.95, and maximum reasoning effort. Judge the result against a concrete task before changing those settings.
Start using GLM 5.3 FlashX in Devin
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into Devin:
set up https://spiral.ai/SKILL.md


