How to write prompts for MOSS Image
MOSS Image makes pictures from a sentence and is built to put text inside the picture, in Chinese and English. This guide rests on the authors’ own user guide and model card, put in our words. We have no sample picture of our own with this model to quote yet, so it states nothing as our test.
Written from the provider’s documentation alone, with no result of ours to quote. Last checked November 8, 2026.
Language, formats and negative tags
From the provider’s documentation
- The authors’ user guide recommends writing in English, which has the most training data, and says Chinese and other languages are supported. The model takes natural-language sentences, tag sequences and JSON layouts.
- The user guide is written for the Base version of the model. The model on our shelf is the SFT version, fine-tuned from it with the same design. We did not check which advice is meant for which, so treat it as general advice.
- For tag sequences the authors suggest an order: quality, score, meta, year, character, series, artist, then general tags.
- For a negative prompt the authors name a few short tags that worked in their tests, such as worst quality, bad anatomy, watermark and jpeg artifact. Pickshot passes a negative prompt to this model when you write one.
- The authors recommend a total of about one, one and a half or two million pixels, and say the model can stretch from a quarter of a million up to four million. Sizes must be multiples of 16.
Text in the picture, and layout
From the provider’s documentation
- For text, put the wording in quotation marks inside the prompt. The model card’s own example describes a kitchen scene and gives a handwritten Chinese note on the fridge door in quotes.
- The model can place text and break lines by itself, but the user guide says the result may not always match what you expect. To get the lines you want, put the line breaks into the prompt yourself, one line of text per line.
- The authors name strong text rendering and layout control as the model’s features, and their examples show Chinese and English text, including a long handwritten Chinese board. They give no separate rules for Chinese and for English text.
- A JSON layout, which the authors mark as experimental, places text and objects by boxes. We have not tried it, and we do not know whether the Pickshot prompt box passes such a prompt to the model unchanged.
Settings of this model
- Kuanzia
- OpenMOSS
- Aina
- AI image model
- Jinsi ya kuunda
- Maandishi kuwa picha
- Ubora
- 1K / 2K
- Bei
- from 84 / image
Maswali
Does MOSS Image write Chinese text well?
Its authors present text in Chinese and English as a main feature and show Chinese examples, and they say the result may not always match what you expect. We have no sample of our own, so check the spelling of every picture.
Which language should I write the prompt in?
The authors recommend English, which has the most training data, and say Chinese and other languages are supported. We did not test this ourselves.
How do I control where the text breaks into lines?
Put the line breaks into the prompt yourself, one line of text per line, as the authors advise. We have not tested how reliably this works on Pickshot.
Provider documentation we read
We read these pages and wrote the points that matter in our own words. The provider may change them.
- MOSS-Image user guide (OpenMOSS, GitHub) (github.com)
- MOSS-Image-SFT model card (OpenMOSS, Hugging Face) (huggingface.co)
More prompting guides
- How to write prompts for GPT Image 2.6 FlareAI image model
- How to write prompts for GPT Image 2.6 SunburstAI image model
- How to write prompts for Nano Banana 2.1AI image model
- How to write prompts for Grok Imagine ProAI image model
- How to write prompts for Nano Banana ProAI image model
- How to write prompts for Nano Banana 2AI image model
- How to write prompts for GPT Image 2.5 FlareAI image model
- How to write prompts for IndigoAI image model
- How to write prompts for SaffronAI image model
- How to write prompts for CeladonAI image model
- How to write prompts for LapisAI image model
- How to write edit requests for UmberKihariri cha picha cha AI
- How to write prompts for Kling Video 4.0AI video model
- How to write prompts for Seedance 2.5AI video model
- How to write prompts for Seedance 2.0 MiniAI video model
- How to write prompts for Kling V3AI video model
- How to write prompts for Kling O3 OmniAI video model
- How to write prompts for Veo 3.1AI video model
- How to write prompts for Veo 3.1 FastAI video model
- How to write prompts for SilkAI video model
- How to write prompts for LacquerAI video model
- How to write prompts for SatinAI video model
Majina ya modeli, kampuni na bidhaa kwenye tovuti hii ni alama za biashara za wamiliki wao; Pickshot haihusiani nao, haifadhiliwi nao wala haiidhinishwi nao.