Bring Your Own Keys, Then Let the Panel Decide Which Model Answers
Connect the provider keys you already pay for, seat several models in one chat, and let Judge mode surface the answer that holds up.
TL;DR
With bring-your-own-keys (BYOK), you connect the provider API keys you already pay for and the same models appear inside aiDex, billed through your own accounts. Seat several in one conversation, then use Compare or Judge mode so the panel surfaces the answer that holds up instead of betting on one model. You can mix your own keys with managed credits or local Ollama models, and per-message costs stay in view the whole time.
Most multi-model tools make you rent their credits. aiDex lets you bring your own: connect the API keys you already pay for, and the same OpenAI, Anthropic, Google, and DeepSeek models show up in your chat. Then, instead of betting on one model, you put the question to several and let the panel surface the answer that holds up.
That is the whole idea behind a bring-your-own-keys panel. You keep the billing relationship with each provider, and aiDex handles the orchestration: seating the models in one conversation, running the speaking order, and showing the per-message cost as you go.
What does "bring your own keys" mean in aiDex?
It means you connect your own provider API keys, and those models answer inside aiDex billed through your provider account. BYOK (bring your own keys) is the alternative to running on aiDex's managed credits, and you can mix the two: some models on your keys, some on ours.
aiDex seats up to five models in one conversation (Claude Opus 4.8, GPT-5.4, Gemini 3.1 Pro, DeepSeek V3.2, plus local Ollama models). With BYOK the cloud models run on the keys you add; with managed credits they run on a balance you top up here. Use your own provider keys or the ones we manage, and pick the models you want.
How do I add my own provider keys?
Open settings, paste the API key for each provider you use, and the matching models appear in the Dex, aiDex's model catalog. You add a key once per provider (OpenAI, Anthropic, Google, DeepSeek), and every model that key unlocks becomes selectable in any chat.
From there you build a panel: open Teams, pick the models you want at the table, and ask. The keys stay yours, the conversation stays in one place, and you can swap a model in or out between messages without leaving the chat.
How do I let the panel decide which model answers?
Put the same question to several models, then use Judge mode to have one model weigh the answers and pick. The flow is short:
- Ask the panel. In Team or Compare, send your question to two or three models at once. Each answers independently.
- Let a model judge. Judge mode hands the answers to one model with a brief: weigh these, say where they agree and disagree, and recommend the strongest.
- You sign off. The Judge narrows the field; the final call is yours.
If you would rather pick yourself, stop after Compare and read the answers side by side. For the consensus pattern in depth, see how to get a consensus answer from several AIs.
Compare or Judge: who picks the winning answer?
Use Compare when you want to choose yourself; use Judge when you want a model to choose first. Compare lays the answers out in columns so you do the weighing. Judge adds a model that reconciles them and recommends one, with your sign-off.
For a question where you know the answer when you see it, Compare is faster. For a stack of questions where you want a first pass that flags only the splits, Judge saves the most time. To map every mode to a situation, see when to use each aiDex mode.
What does BYOK cost compared with managed credits?
With BYOK you pay each provider directly at their rates; with managed credits you pay aiDex from a balance you top up. Either way, aiDex shows the per-message cost in the chat and lets you set spending limits, so a five-model panel never surprises you.
The practical move is to size each seat to its job: a flagship model for the hard read, a cheaper model for the second opinion, and a light model in the Judge seat, since weighing answers is lighter work than producing them. To choose by criteria rather than reputation, see which AI model for which task.
Can I mix my own keys with local Ollama models?
Yes. One panel can hold cloud models on your keys and local Ollama models on your own machine in the same conversation. That lets a private local model handle sensitive context while a frontier cloud model handles the heavy reasoning, all in one chat. For the local setup, see bringing Ollama into your aiDex chat.
Your next hard question does not need you to guess the right model in advance. Add your keys once, seat a panel, and let Judge mode surface the answer that survives a second look. For the bigger picture, start with the guide to multi-model AI workflows.
The aiDex Team · Multi-model AI platform
aiDex is a multi-model AI platform that lets you query several AI models at once, compare their answers, run consensus picks, and chain models in pipelines or open team chats. Use your own provider keys or the ones we manage, and pick the models you want.
Frequently asked questions
What does bring your own keys (BYOK) mean in aiDex?
BYOK means you connect your own provider API keys, so OpenAI, Anthropic, Google, and DeepSeek models answer inside aiDex billed through your own provider accounts. It is the alternative to aiDex's managed credits, and you can mix the two in one chat.
How do I add my own API keys?
Open settings in aiDex and paste the API key for each provider you use. The models that key unlocks then appear in the model catalog and become selectable in any conversation. You add a key once per provider.
How do I let the models decide which answer is best?
Send the same question to several models in Team or Compare, then use Judge mode to have one model weigh the answers and recommend the strongest. You keep the final sign-off; the Judge just narrows the field to the points that matter.
Is BYOK cheaper than managed credits?
It depends on your volume. With BYOK you pay each provider directly at their rates; with managed credits you pay aiDex from a balance. Use your own provider keys or the ones we manage, and pick the models you want.
Can I mix my own keys with local Ollama models?
Yes. One panel can hold cloud models on your keys and local Ollama models on your own machine in the same conversation, so a private local model can handle sensitive context while a cloud model does the heavy reasoning.
Keep reading
Multi-Model AI Workflows: Why Query All Models at Once (2026 Guide)
One model is one opinion. Here is how to query several at once and get a better answer.
How to Create a Multi-AI Team in aiDex
Build a panel of named AI personas, each pinned to its own model, with a moderator that watches for consensus.
Bring Ollama Into Your aiDex Chat: Local Models, Same Table
Run open-weight models on your own machine and mix them with cloud models in one conversation.
Which AI Model for Which Task? A Practical 2026 Routing Guide
Match the model type to the job, then compare 2 to 3 candidates on your real prompt instead of guessing.