SLATE — free OSS prompt generator for image/video/music models

SLATE: a free, open-source prompt generator that writes optimised prompts for image/video/music models, wired to local models or existing subscriptions — aimed at creators, not developers.

Updated 2026-08-09 · Source https://x.com/SamJWasserman/status/2085469672489799752 · 中文版

What is the idea?

Open-source tool that turns plain descriptions into model-tuned prompts across image/video/music backends, local or subscription.

What is missing today?

Only a shrinking one: a Chinese-model version (即梦/可灵/海螺) doesn't exist — but the underlying pain is being absorbed by the models themselves.

What makes this one different?

If touched at all: the Chinese-model adaptation, sold as a workflow inside 即梦/可灵 communities — but the honest read is the window is closing.

What could kill it?

Prompt optimisation is a feature every generation platform ships natively, and each model generation needs it less. A band-aid on a closing gap — a feature, not a company.

Who pays, and for what?

Creators on Chinese video/image models, one-time or freemium.

What is the main risk?

Model progress deletes the job it does.

How much work is it?

about two weeks of work

Where does the evidence come from?

https://x.com/SamJWasserman/status/2085469672489799752

XArchivedTwo weeks of workx-daily

Paygent — virtual cards for AI agents to pay autonomously…  ·  Soloop — approval-gated 'agent OS' for solo founders (PH #2)