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

# Seed 2.1 Turbo

> Production-grade multimodal reasoning model for coding, long-horizon agent execution, video understanding, and structured output.

Seed 2.1 Turbo is BytePlus ModelArk's next-generation model for coding and agent workflows, with stronger engineering-grade code delivery, autonomous planning, and dynamic self-repair.

It supports text generation and multimodal understanding through both Chat Completions and Responses APIs, including video understanding, function calling, and structured output.

Use it for long-running production tasks that need deep reasoning, tool use, and visual context within a 256K-token context window.

## Model ID

```text theme={"dark"}
dola-seed-2-1-turbo-260628
```

Pass this exact public ID in the top-level `model` field. Do not substitute an upstream provider alias.

## Supported modes

* `responses_api`
* `visual_understanding`
* `text_generation`
* `chat_completions`

## Inputs

* `Text`
* `Image`
* `Video`

## Outputs

* `Text`

## Capabilities

* `Visual understanding`
* `Text generation`

### Additional capabilities

* `Online inference`
* `Context caching`
* `Deep reasoning`
* `Function calling`
* `Structured output`

## Technical details

| Property             | Value                             |
| -------------------- | --------------------------------- |
| Provider             | `byteplus-modelark`               |
| Modality             | `reasoning`                       |
| Processing           | `Synchronous response`            |
| Endpoint             | `/api/gengen/v1/chat/completions` |
| Rate limit           | `500`                             |
| Context window       | `256K`                            |
| Max input tokens     | `256K`                            |
| Max output tokens    | `256K`                            |
| Max reasoning tokens | `256K`                            |
| Token rate limit     | `1000K`                           |

<Note>
  Availability and final customer prices can change. Check the [Dashboard model catalog](https://gengen.farm/dashboard?section=models) before production use.
</Note>

See the [API reference overview](/api-reference/introduction) for authentication, request conventions, and response schemas.