---
title: Hosted models
description: Open-weight models Lovely Router runs on its own dedicated GPUs, in the Creative · Hosted group. They sleep when idle, so the first request after a quiet spell takes a little longer.
---

Some open-weight models aren't offered by any API provider, or not without a content filter. Lovely Router runs these on its own GPUs and serves them in the `creative-hosted` channel group. You call them like any other model: same keys, same balance, same endpoints.

## Available models

| Model id | What it is | Context | Input | APIs |
| --- | --- | --- | --- | --- |
| `dongfangshuo-27b-v2` | Dongfangshuo (东方槊) 27B v2: Qwen3.8-27B fine-tuned for Chinese long-form fiction and roleplay, without the base model's refusals. For adults (18+). | 128K tokens | text, images | OpenAI Chat Completions, Anthropic Messages |
| `z-image-turbo-nsfw-v2` | Z-Image Turbo NSFW v2: Z-Image Turbo fine-tuned for adult photographic images, 9 steps. Text-to-image. For adults (18+). | – | text | OpenAI Images |
| `moss-image-sft` | MOSS-Image SFT (OpenMOSS): a 5B text-to-image model, strong at text in images, layouts and Chinese or English prompts. 50 steps. | – | text | OpenAI Images |
| `moss-image-nsfw` | MOSS-Image NSFW: MOSS-Image SFT with an adult-photography LoRA (scale 0.7, 30 steps; its trigger word is added for you). Text-to-image. For adults (18+). | – | text | OpenAI Images |

Prices are on the [marketplace](/models) and in [`GET /v1/models`](/docs/models/pricing#list-models).

The video models also run on our own GPUs, each with its own guide: [character swap](/docs/models/videos), [live video](/docs/models/live-video) and [music videos](/docs/models/music-videos). So do the speech models, `qwen3-tts` and `zonos2`: see [Text to speech](/docs/models/speech).

## The first request after an idle spell {#cold-start}

A hosted model sleeps when nobody has used it for 2 minutes. The next request wakes it, and waits while it does: usually **10–30 seconds** before the first token, occasionally up to a couple of minutes. After that it answers at normal speed, and it stays awake while requests keep coming.

You don't need to retry: the gateway holds your request until the model is up. Give your client a timeout of at least **3 minutes** for the first call. If the model can't wake within about 2½ minutes, you get a `502` you can retry.

Each hosted model runs on a main set of GPUs and a standby set. When no main GPU is free within about 35 seconds, the request moves to the standby set, which also wakes in seconds. A request that fails is not moved to another provider the way it is in the other groups.

## Images {#images}

`z-image-turbo-nsfw-v2` takes the same generation requests as the other [image models](/docs/models/images), at its own flat price per image:

- `size`: `1024x1024`, `1536x1024`, `1024x1536`, `1536x1536`, or any `WIDTHxHEIGHT` up to 1536 pixels on the long side, multiples of 16. `auto` or no size gives `1024x1024`. Larger sizes are refused with `400`.
- A warm model returns a 1024×1024 image in about 5 seconds, 1536×1536 in about 15.
- `quality` has no effect. Every image uses a new random seed.
- It generates only: `/v1/images/edits` and reference images return `400`.

`moss-image-sft` and `moss-image-nsfw` take the same generation requests at sizes up to **1024 pixels** on the long side (multiples of 16; `auto` gives `1024x1024`), at their own flat price per image. They are slower: a warm model takes about **50 seconds** for a `moss-image-sft` image and about **30 seconds** for `moss-image-nsfw`. They generate only.

With [`n`](/docs/models/images#several-images), a hosted model makes the images one after another on the GPU that is already awake, so they add up: four `z-image-turbo-nsfw-v2` images at 1024×1024 take about 20 seconds once the model is up, two `moss-image-nsfw` images about a minute.

## Thinking

`dongfangshuo-27b-v2` answers directly by default, which is what its authors recommend for fiction. To let it reason first, turn thinking on per request:

```json title="Chat Completions body"
{
  "model": "dongfangshuo-27b-v2",
  "messages": [{"role": "user", "content": "…"}],
  "chat_template_kwargs": {"enable_thinking": true}
}
```

The reasoning comes back in `message.reasoning_content` (streams: `delta.reasoning_content`), separate from the answer.

## Sampling

Defaults: `temperature` 0.8, `top_p` 0.9, `top_k` 50, `repeat_penalty` 1.15 (tuned by the model's authors against repetition in long passages). Send any of them to override. Besides the OpenAI parameters, Chat Completions also accepts `top_k`, `min_p` and `repeat_penalty`.

For long fiction, ask for about 2,000–4,000 Chinese characters per reply and continue chapter by chapter; a structured prompt (scene, characters and their state, previous chapter, plot steps, required details) gives the model the most control.

## Content

These models follow the prompt without the refusals of their base models. You are responsible for what you generate with them: fiction for adults only. The [Acceptable Use Policy](/acceptable-use) applies as for every model; sexual content involving minors is never allowed.