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

# Models

> Quick guide to picking the right model for your use case.

<Tip>
  **Not sure which model to pick?** Set `model: "auto"` and NinjaChat picks the best one for your prompt. [Learn more about smart routing](/smart-routing)
</Tip>

## Pick a model

Chat prices below are a **typical price** — a derived reference at 5K input / 1K output tokens, not a flat fee. Chat bills per token; see [Pricing](/pricing) for the live rates.

| I want to...                      | Model                                                             | Typical price         | ID                  |
| --------------------------------- | ----------------------------------------------------------------- | --------------------- | ------------------- |
| Let NinjaChat decide              | Auto (Balanced)                                                   | resolved model's rate | `auto`              |
| Build a production chatbot        | [GPT-5](/api-reference/models/chat/gpt-5)                         | \~\$0.0255            | `gpt-5`             |
| Generate high-quality code        | [Claude Sonnet 4.6](/api-reference/models/chat/claude-sonnet-4-6) | \~\$0.0336            | `claude-sonnet-4.6` |
| Do math and logical reasoning     | [o3-mini](/api-reference/models/chat/o3-mini)                     | \~\$0.0112            | `o3-mini`           |
| Get the best possible answer      | [Claude Opus 4.6](/api-reference/models/chat/claude-opus-4-6)     | \~\$0.056             | `claude-opus-4.6`   |
| Classify or extract text (budget) | [DeepSeek V3](/api-reference/models/chat/deepseek-v3)             | \~\$0.0064            | `deepseek-v3`       |
| Fast real-time chat               | [Gemini 3 Flash](/api-reference/models/chat/gemini-3-flash)       | \~\$0.0066            | `gemini-3-flash`    |
| Analyze long documents            | [Gemini 2.5 Pro](/api-reference/models/chat/gemini-2-5-pro)       | \~\$0.0182            | `gemini-2.5-pro`    |
| Generate images                   | [Nano Banana 2](/api-reference/models/image/nano-banana-2)        | \$0.06/img            | `nano-banana-2`     |
| Generate video                    | [Veo 3.1 Fast](/api-reference/models/video/veo-3-1-fast)          | \$3.00/video          | `veo-3.1-fast`      |
| Search the web                    | [Web Search](/search)                                             | \$0.05/query          | —                   |

Browse all models with full specs, context windows, and pricing in the [Models tab](/models-overview).

## Fetch models via API

```bash theme={null}
GET https://www.ninjachat.ai/api/v1/models
```

Returns metadata per model: `capabilities`, `context_window`, `max_output_tokens`, `speed`, and `recommended_for`. Supports filtering by `tier`, `capability`, and `virtual`.