This part continues the Handmade ChatGPT you built before — its code is
carried into this folder. Setup only; the build is the next task.

Confirm AGENTS.md still declares, each at the start of its own line:

stack: <language, framework, storage>
run:
test:

They should already be there from providers you carried in. Make sure
run: still starts the product and test: still runs a suite that passes —
a part carried honestly starts from working code.

Then plan THIS part in TODO.md: add the steps for a model per conversation, chosen and switched, in
product terms, and check them off as they land. AGENTS.md and TODO.md are
read by every judge of this part as evidence of how you work.

Every conversation runs on a model the person chose. You decide when it
is done; a judge panel scores the result.

Scenarios:

Scenario: Models come from the provider
When a provider is added
Then the models it offers appear in the model picker as "/"
And a provider that lists none shows as empty, not as broken

Scenario: A model per conversation
When someone starts a conversation
Then it opens on the default model, and the picker can change it
And the conversation shows which model it is on

Scenario: Switching mid-way
Given a conversation with two exchanges on one model
When someone picks a model of another provider and sends a message
Then the new provider receives the whole history
And the earlier replies still say which model wrote them

Scenario: A default model
When someone sets a default model in settings
Then new conversations open on it
And existing conversations keep theirs

Everything beyond the scenarios is yours: where the picker sits, how a
model is described, what happens to a conversation whose model is gone.

When you are done, write .ololo/chatgpt-models-done.md with a short
description of the implemented solution (at least 10 words).