Proposal: Add Aurelo as a first-class model provider

[1abhayrajsinghbhadauriya-dev]

Hi Late CLI maintainers,

I’m Abhay Raj Singh Bhadauriya, Founder and CEO of Aurelo.

I’d like to propose adding Aurelo as a first-class model provider in Late CLI, with Aurelo shown by name in the standard provider/model interface and supported models directly selectable by users.

Aurelo offers the M1 and M2 model families through a documented API designed for coding, complex reasoning, and agentic software-development workflows. Documentation covering models, authentication, context and output limits, streaming, reasoning controls, supported inputs, and pricing is available here:

https://aurelo.tech/docs

Late CLI’s explicit support for compatible APIs and hybrid model routing makes a first-class Aurelo preset technically lightweight.

The Aurelo team is prepared to implement and maintain the integration, including:

• Provider authentication and configuration
• M1 and M2 model selection and metadata
• Streaming and error handling
• Reasoning and capability settings
• Tests and documentation
• Ongoing compatibility maintenance

We can also provide evaluation access so the maintainers can test Aurelo in real Late CLI workflows before deciding whether to accept the integration.

Our primary request is native, first-class provider support. If core inclusion is not preferred, we are open to an officially maintained adapter or extension, provided users can still select Aurelo through a supported and documented experience.

Before preparing a pull request, could you please advise:

1. Whether a first-class Aurelo integration is within scope.
2. Which provider interface or extension mechanism we should use.
3. What compatibility tests or acceptance criteria you would expect.
4. Whether you would prefer a technical design proposal before implementation.

Thank you for considering the proposal.

Best regards,

Abhay Raj Singh Bhadauriya
Founder & CEO, Aurelo
Documentation: https://aurelo.tech/docs
Contact: hello@aurelo.tech

[mlhher]

We support the standard OAI API.