switchloom

# Switchloom

> **Experimental. I currently recommend against adopting this routing workflow.**
>
> I built Switchloom to see whether mixing models could get me the same quality
> for less money. My benchmarks did not deliver that result. In the second
> Pokédex run, the team with Jev cost almost as much as Astra alone and hit the
> time limit before final acceptance. Astra finished and passed all 14 independent
> functional checks.
>
> I am keeping the code and results public, but I would not add this layer to my
> everyday workflow based on these results. This is evidence from two pilots on
> one application, not proof that all model routing fails.
> [Read the benchmark and its limitations](https://switchloom.ai/benchmarks/pokedex-round-2).

**One workflow prompt for persistent Codex tasks, with optional TypeSafe/Jev routing.**

Assign capabilities to models, then copy the prompt into Codex Desktop with
your task. Each model uses a persistent task with its own context. Assignments
end the caller's turn; substantive replies resume it. Codex owns execution.

| Default model | Effort | Capabilities |
| --- | --- | --- |
| GPT-5.6 Luna | max | Coordination, mechanical tasks |
| GPT-5.6 Sol | medium | Implementation, debugging, tests & validation |
| GPT-6 Astra | high | Planning, code review, browser, computer use, visual design & review |

Spatial and 3D modeling starts disabled. Move capabilities between cards or into Disabled, using drag-and-drop
or the arrow menu. Only assigned capabilities contribute instructions. Model
cards can be disabled. Re-enabling a card restores its default model, effort
and capabilities, reclaiming those capabilities from any other card. The reset
icon restores the whole board, including Jev on. Settings are session-local.

Choose another su