> ## Documentation Index
> Fetch the complete documentation index at: https://docs.ruapi.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# GLM models: which to choose and what they can do

> How GLM-5.2, GLM-5-Turbo and the GLM-5V vision model differ, which to pick for everyday work, agents, code or reading images, and what's supported: vision, tool use, low price.

GLM is Zhipu AI's model family. Through RuAPI it comes in three lines — a **flagship**, a fast **turbo**, and a **vision** model that also reads images. People reach for GLM for the same reasons: strong agent and tool-use behaviour, solid code and bilingual text (English and Chinese), all at a low price. This page helps you pick a model for the task; the technical side of connecting is in the [Quickstart](/en/quickstart).

## How the lines differ

| Line                      | Strong at                                                  | Speed / cost    | Reach for it when                                 |
| ------------------------- | ---------------------------------------------------------- | --------------- | ------------------------------------------------- |
| **GLM-5.2** (flagship)    | Strongest all-round: reasoning, code, agents, long context | Full power      | The hardest tasks, when you want the best quality |
| **GLM-5-Turbo** (turbo)   | Fast all-rounder: chat, code, general tasks                | Faster, cheaper | Everyday work and high volume                     |
| **GLM-5V-Turbo** (vision) | Understands images **plus** text                           | Fast            | Reading screenshots, photos, diagrams, charts     |

<Tip>
  Not sure? Start with GLM-5-Turbo — it covers most work fast and cheap. Move up to GLM-5.2 when the task needs the best quality, and reach for GLM-5V-Turbo only when you actually pass an image.
</Tip>

## What GLM can do

* **Vision** — GLM-5V-Turbo takes images as input: screenshots, photos, diagrams, charts. The text-only lines don't. See [vision](/en/vision) for how to pass an image.
* **Agents and tool use** — GLM is strong at agentic workflows and returns structured tool calls reliably; see [function calling](/en/function-calling) for a full example.
* **Strong code and reasoning** — the flagship writes and edits code confidently and handles step-by-step reasoning.
* **Bilingual** — particularly strong on Chinese alongside English, useful for mixed-language content.
* **Low price** — GLM's key draw: capable models at a noticeably lower per-token cost.

For long answers, you can also [stream the response](/en/streaming) token by token instead of waiting for the full reply.

## Picking by task

* **Best quality on a hard task** → GLM-5.2.
* **Cheap and high volume** → GLM-5-Turbo, for price and speed.
* **Reading images** → GLM-5V-Turbo (vision).
* **Agent or tool pipeline** → GLM-5.2 or GLM-5-Turbo (both have reliable tool use).
* **Code and refactoring** → GLM-5.2 for hard work, GLM-5-Turbo for everyday.

## How to connect

GLM is served over the **OpenAI-compatible** protocol — no separate library needed, take any OpenAI SDK and change the `base_url` to `https://www.ruapi.ai/v1`. Step-by-step setup and a first request are in the [Quickstart](/en/quickstart).

<Note>
  The `model` field takes an ID, e.g. `GLM-5.2` for the flagship, `GLM-5-Turbo` for the fast all-rounder and `GLM-5V-Turbo` for vision. Exact names, versions and per-token prices are on the **Pricing** page at [www.ruapi.ai](https://www.ruapi.ai).
</Note>

## FAQ

<AccordionGroup>
  <Accordion title="Which GLM model should I pick?">
    GLM-5-Turbo is the everyday pick — fast, cheap, covers most work. Use GLM-5.2 when you need the best quality on a hard task, and GLM-5V-Turbo when you need to pass an image.
  </Accordion>

  <Accordion title="Does GLM support images?">
    Yes — GLM-5V-Turbo is a vision model and takes images as input. The other GLM lines are text-only. For more on passing images, see [vision](/en/vision).
  </Accordion>

  <Accordion title="Is GLM good for agents and tool use?">
    Yes. GLM is strong at agentic workflows and returns structured tool calls reliably; a full example is in [function calling](/en/function-calling).
  </Accordion>

  <Accordion title="How much does it cost?">
    GLM is one of the better-value families per token: capable models at a noticeably lower price. Exact prices are on the main site's Pricing page.
  </Accordion>

  <Accordion title="What context length does GLM have?">
    Large — the window fits long documents and big chunks of code. The exact limit depends on the version and matches Zhipu's official one.
  </Accordion>

  <Accordion title="Where do I get a key, and how do I pay?">
    You create a key in your dashboard after signing up; wiring it into code is in the [Quickstart](/en/quickstart). Top up in USDT — see the [Top-up](/en/topup) page.
  </Accordion>
</AccordionGroup>

## Next

<CardGroup cols={2}>
  <Card title="Connect and first request" icon="bolt" href="/en/quickstart">
    base\_url, key, Python and curl code.
  </Card>

  <Card title="Vision: reading images" icon="image" href="/en/vision">
    Pass an image to GLM-5V-Turbo.
  </Card>

  <Card title="DeepSeek models" icon="microchip" href="/en/deepseek-api">
    Same key — another strong value line.
  </Card>
</CardGroup>