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

# Model List

> The current Nito catalog, grouped by privacy level, with context window, price, and capabilities.

<Note>
  Snapshot of **2026-08-13**. Run [`models`](/commands/choosing-a-model) for the live list on your account. **Price /1M** is USD per one million tokens, shown as **input · output**. The **privacy level** is shown as its own section, because [the level follows the model you choose](/privacy/levels). See [Pricing](/models/pricing) for how cost is structured.
</Note>

Capabilities in the table:

* **reasoning** step-by-step thinking. See [Reasoning](/features/reasoning).
* **vision** image input. See [Image Input](/features/multimodal/image-input).
* **web** web search and fetch. See [Web Search](/features/server-side-tools/web-search).
* **attestable** supports a verifiable attestation. See [The Four Levels](/privacy/levels).

## Anonymous (16 models)

Frontier models. The provider gets your request from Nito, without your identity attached. See [Anonymous](/privacy/levels/anonymous).

| No. | Model ID                        | Context |     Price /1M    | Capabilities           |
| :-: | :------------------------------ | :-----: | :--------------: | :--------------------- |
|  1  | `anthropic/claude-haiku-4-5`    |   200k  |  \$1.00 · \$5.00 | reasoning, vision, web |
|  2  | `anthropic/claude-opus-4.8`     |  1000k  | \$5.00 · \$25.00 | reasoning, vision, web |
|  3  | `anthropic/claude-opus-5`       |  1000k  | \$5.00 · \$25.00 | reasoning, vision, web |
|  4  | `anthropic/claude-sonnet-5`     |  1000k  | \$2.00 · \$10.00 | reasoning, vision, web |
|  5  | `google/gemini-3.1-flash-lite`  |  1049k  |  \$0.25 · \$1.50 | reasoning, vision, web |
|  6  | `google/gemini-3.1-pro-preview` |  1049k  | \$2.00 · \$12.00 | reasoning, vision, web |
|  7  | `google/gemini-3.5-flash`       |  1049k  |  \$1.50 · \$9.00 | reasoning, vision, web |
|  8  | `openai/gpt-5.3-codex`          |   400k  | \$1.75 · \$14.00 | reasoning, vision, web |
|  9  | `openai/gpt-5.4-mini`           |   400k  |  \$0.75 · \$4.50 | reasoning, vision, web |
|  10 | `openai/gpt-5.5`                |  1050k  | \$5.00 · \$30.00 | reasoning, vision, web |
|  11 | `openai/gpt-5.6-luna`           |  1050k  |  \$0.10 · \$0.60 | reasoning, vision, web |
|  12 | `openai/gpt-5.6-luna-pro`       |  1050k  |  \$0.10 · \$0.60 | reasoning, vision, web |
|  13 | `openai/gpt-5.6-sol`            |  1050k  | \$5.00 · \$30.00 | reasoning, vision, web |
|  14 | `openai/gpt-5.6-sol-pro`        |  1050k  | \$5.00 · \$30.00 | reasoning, vision, web |
|  15 | `openai/gpt-5.6-terra`          |  1050k  |  \$1.00 · \$6.00 | reasoning, vision, web |
|  16 | `openai/gpt-5.6-terra-pro`      |  1050k  |  \$1.00 · \$6.00 | reasoning, vision, web |

## Private (9 models)

Everything Anonymous gives you, plus the upstream provider held to zero data retention. See [Private](/privacy/levels/private).

| No. | Model ID                            | Context |     Price /1M    | Capabilities           |
| :-: | :---------------------------------- | :-----: | :--------------: | :--------------------- |
|  1  | `minimax/minimax-m3`                |  1049k  |  \$0.30 · \$1.20 | reasoning, vision, web |
|  2  | `moonshotai/kimi-k3`                |  1049k  | \$3.00 · \$15.00 | reasoning, vision, web |
|  3  | `nousresearch/hermes-4-405b`        |   131k  |  \$1.00 · \$3.00 | reasoning, web         |
|  4  | `nvidia/nemotron-3-ultra-550b-a55b` |   512k  |  \$0.60 · \$3.60 | reasoning, web         |
|  5  | `qwen/qwen3-235b-a22b-2507`         |   262k  |  \$0.09 · \$0.55 | reasoning, web         |
|  6  | `qwen/qwen3-coder`                  |   262k  |  \$0.30 · \$1.00 | reasoning, web         |
|  7  | `qwen/qwen3-vl-235b-a22b-instruct`  |   262k  |  \$0.26 · \$1.04 | reasoning, vision, web |
|  8  | `x-ai/grok-4.6`                     |   500k  |  \$2.00 · \$6.00 | reasoning, vision, web |
|  9  | `z-ai/glm-5.2`                      |  1049k  |  \$0.63 · \$1.98 | reasoning, web         |

## Confidential (9 models)

Sealed execution with a verifiable attestation. Listed with a `:confidential` marker on the model name. See [Confidential](/privacy/levels/confidential).

| No. | Model ID                                         | Context |    Price /1M    | Capabilities                       |
| :-: | :----------------------------------------------- | :-----: | :-------------: | :--------------------------------- |
|  1  | `deepseek-ai/deepseek-v4-flash:confidential`     |  1049k  | \$0.17 · \$0.35 | reasoning, web, attestable         |
|  2  | `google/gemma-4-26b-a4b-uncensored:confidential` |   66k   | \$0.15 · \$0.70 | vision, web, attestable            |
|  3  | `google/gemma-4-31b-it:confidential`             |   262k  | \$0.13 · \$0.40 | reasoning, vision, web, attestable |
|  4  | `moonshotai/kimi-k2.6:confidential`              |   262k  | \$0.81 · \$3.85 | reasoning, vision, web, attestable |
|  5  | `openai/gpt-oss-120b:confidential`               |   131k  | \$0.15 · \$0.55 | reasoning, web, attestable         |
|  6  | `qwen/qwen3-vl-30b-a3b-instruct:confidential`    |   16k   | \$0.15 · \$0.55 | reasoning, vision, web, attestable |
|  7  | `qwen/qwen3.5-397b-a17b:confidential`            |   128k  | \$0.50 · \$3.30 | reasoning, web, attestable         |
|  8  | `qwen/qwen3.6-35b-a3b-uncensored:confidential`   |   131k  | \$0.30 · \$1.50 | vision, web, attestable            |
|  9  | `z-ai/glm-5.2:confidential`                      |  1049k  | \$1.40 · \$4.40 | reasoning, web, attestable         |

## Encrypted (coming soon)

Client-side encryption to the sealed environment, the strongest level. The plugin and the `nito` command line tool cannot run a call at this level in this release, so the catalog you browse returns no Encrypted entries and an `:encrypted` model is refused. Every model that will carry it is a model already listed under Confidential above. See [Encrypted](/privacy/levels/encrypted).

## Using a Model From the List

Set one as your default, or use it for a single call:

<Tabs>
  <Tab title="Claude Code">
    ```text theme={"theme":{"light":"github-light","dark":"github-dark"},"languages":{"custom":["typescript","python","curl"]}}
    /nito:model z-ai/glm-5.2
    /nito:ask --model qwen/qwen3-235b-a22b-2507 Summarize this thread.
    ```
  </Tab>

  <Tab title="Codex">
    ```text theme={"theme":{"light":"github-light","dark":"github-dark"},"languages":{"custom":["typescript","python","curl"]}}
    $nito:model z-ai/glm-5.2
    $nito:ask --model qwen/qwen3-235b-a22b-2507 Summarize this thread.
    ```
  </Tab>
</Tabs>

Not sure which to pick? See [Choosing the Right Model](/models/choosing-a-model). For how model IDs are named, see [Model ID Convention](/models/model-id-convention). To set defaults and switch models, see [Choosing a Model](/commands/choosing-a-model).