# GLM prompt guide: constraints early, useful roles and explicit JSON

> How to write prompts for GLM-5.3 and GLM-5.3 Flash from Z.ai (Zhipu AI), following Zhipu's general prompt-engineering guide and the GLM-5.3 docs.

Source: Promptea — https://www.promptea.me/en/guides/glm-prompt-guide · Language: English · Last updated: 2026-11-14

## What Zhipu recommends

- Give the important constraints you already know as early as possible. A rule that arrives at the end is easy to under-weight.
- Roles should serve the task. Skip inflated identities like 'world-class expert'; a role cannot replace the task description, the source evidence or the validation criteria.
- Name tags after what they contain and keep them consistent, so the model can tell the requirements it must follow from the material it must process.
- The number of examples is not the goal: cover the differences that matter.
- For JSON, spell out the structure, what each field means and the allowed values. A structured prompt describes the result but cannot guarantee valid data, so validate it in your app.

## What is different about GLM-5.3

- Reasoning is always on: disabling it is no longer supported. Pick reasoning_effort low, high or max (default max); low is the setting for quick tasks.
- GLM-5.3 takes text only, with a 1M-token context and up to 128K tokens of output. Use GLM-5.3 Flash for images, video and files.
- Because it already reasons, you do not need 'think step by step'. Spend the words on constraints and on how to check the answer.

## Templates

### Constraints-first review

```text
Constraints (apply to everything below):
- Audience: [who]
- Max length: [limit]
- Do not: [what to avoid]

Task: review the draft below and return the 5 most important fixes, most important first.

<draft>
[paste the draft]
</draft>

Done when: each fix names the sentence it applies to and the reason.
```

[Open in Promptea](https://www.promptea.me/en/analyzer?prompt=Constraints+%28apply+to+everything+below%29%3A%0A-+Audience%3A+%5Bwho%5D%0A-+Max+length%3A+%5Blimit%5D%0A-+Do+not%3A+%5Bwhat+to+avoid%5D%0A%0ATask%3A+review+the+draft+below+and+return+the+5+most+important+fixes%2C+most+important+first.%0A%0A%3Cdraft%3E%0A%5Bpaste+the+draft%5D%0A%3C%2Fdraft%3E%0A%0ADone+when%3A+each+fix+names+the+sentence+it+applies+to+and+the+reason.&purpose=text&target=glm)

### JSON with allowed values

```text
Read the review below and return JSON only:
{"sentiment": "positive" | "neutral" | "negative", "topics": string[] (max 3, lowercase), "refund_requested": boolean}
- sentiment: the reviewer's overall view.
- topics: what the review is about (e.g. "delivery", "price").
- refund_requested: true only if the reviewer asks for money back.

<review>
[paste the review]
</review>
```

[Open in Promptea](https://www.promptea.me/en/analyzer?prompt=Read+the+review+below+and+return+JSON+only%3A%0A%7B%22sentiment%22%3A+%22positive%22+%7C+%22neutral%22+%7C+%22negative%22%2C+%22topics%22%3A+string%5B%5D+%28max+3%2C+lowercase%29%2C+%22refund_requested%22%3A+boolean%7D%0A-+sentiment%3A+the+reviewer%27s+overall+view.%0A-+topics%3A+what+the+review+is+about+%28e.g.+%22delivery%22%2C+%22price%22%29.%0A-+refund_requested%3A+true+only+if+the+reviewer+asks+for+money+back.%0A%0A%3Creview%3E%0A%5Bpaste+the+review%5D%0A%3C%2Freview%3E&purpose=data&target=glm)

## Frequently asked questions

### Should I ask GLM-5.3 to think step by step?

No need: it always reasons. Adjust depth with reasoning_effort and use the prompt for constraints, material and success criteria.

### Z.ai or Zhipu AI?

Both are Zhipu AI platforms: Z.ai (docs.z.ai) for international users and BigModel (bigmodel.cn) in China. Promptea lists the model ids from the Z.ai docs.
