Get the most out of dequid —
one job at a time.
Five playbooks for the things people actually use dequid for. Plus a short reference.
Playbooks
Start here
Build your first persona
A worked example you can copy.
Read →Move your AI life into dequid
From one ChatGPT bag to a separate setup per role.
Read →Pick web or Mac
Same product, two data paths.
Read →SEEK recipes→
Four-dial presets you can copy for code review, coaching, research, more.Memory and Context Bleed→
Why ChatGPT memory leaks and how dequid prevents it.How dequid reads your context→
Why the system prompt stays small. How the model fetches your vault, memory, and chats — only when the question warrants it.Keyboard shortcuts→
Every shortcut in the app: shell, composer, sidebar, and search.Import from ChatGPT or Claude→
Settings → Data → Import Conversations → upload .zip or .json. Memories extracted automatically.FAQ
Common questions.
What makes dequid different from ChatGPT memory?+
ChatGPT, Claude, and Gemini all keep one bag of facts about you and use all of it, every question. With dequid, you make a separate persona for each kind of conversation — its own people, documents, memory, and voice. Switch personas, and the entire context the model sees changes. They store. We separate.
Is the web app free?+
Yes. You get starter credits so you can chat with personas, the vault, and memory at your own pace. No card required. Move to Pro when you want ~690K credits included every month, a better rate beyond that, and more storage.
What models can I use?+
Any major LLM — Claude, GPT-4o, Gemini 2.5 Pro, Llama 3.3, DeepSeek R1, Mistral, and many more. The model list updates live.
How is pricing calculated?+
You pay the LLM service provider's per-token rate plus a small dequid markup. Free accounts start with credits on the house. Pro includes ~690K credits every month and a better rate on anything beyond that. Full plan details →