[Feature]: Add Qwen (Tongyi Lingma) as a new provider

[Tvardovsky]

Before submitting

• I searched existing issues and did not find a duplicate.
• I am describing a concrete problem or use case, not just a vague idea.

Area

apps/server

Problem or use case

I want to use Qwen (Tongyi Lingma) as a coding provider in T3 Code. Currently, the application only supports Codex and Claude Code.

Qwen offers compelling advantages for developers:

• 1 million token context window — enables handling massive codebases and maintaining context across long conversations
• Coder subscription at $50/month with generous usage limits — cost-effective for individuals and teams
• Open-source application and models — ensures transparency and community-driven improvements
• Strong coding performance with excellent benchmark results
• Generous free tier — lowers the barrier to entry for testing

Proposed solution

Add Qwen provider support as an additional option alongside Codex and Claude Code. This would involve extending the provider architecture to support Qwen's API and event streaming.

This would give users more provider choices and leverage Qwen's unique strengths (large context window, open-source nature, competitive pricing) for complex coding tasks. They already have an open-source Qwen Code CLI agent, and the t3code will reuse the OAUTH token from it and will require it as a dependency.

Why this matters

Diversifying provider options would:

• Support users who prefer open-source solutions
• Provide cost-effective alternatives for teams with specific needs
• Leverage Qwen's 1M token context for large-scale codebase work
• Improve platform resilience through provider diversity

Smallest useful scope

A minimal implementation would add basic Qwen provider support with session management, core event streaming, and basic error handling.

Alternatives considered

Users currently must choose between Codex and Claude Code only. If you prefer Qwen's capabilities, pricing, or open-source philosophy, you won't be able to use T3 Code with your preferred provider.

Risks or tradeoffs

• Additional maintenance burden for a new provider
• Need to keep up with Qwen API changes
• Potential complexity in provider management logic
• Different behavioral characteristics across providers

Examples or references

Qwen official site: https://qwenlm.github.io/
Qwen Code (open-source): https://github.com/QwenLM/qwen-code

Contribution

• I would be open to helping implement this.