How Magicdoor Auto Chooses DeepSeek, Luna, Sol, or Opus

Direct answer: Magicdoor's Experimental Auto setting chooses before every message. Routine compatible requests use DeepSeek V4.1 Flash; substantive compatible text can use Claude Opus 5.5. OpenAI-required work, including code/data execution, PDFs, or non-image files in the effective conversation, can use GPT-6 Luna when compatible and routine, or GPT-6.1 Sol when Jev signals substantive quality or code/data execution. The previous answer model does not lock the next turn, but saved file and container history can constrain it.

Fact-checked September 23, 2026. This page owns Magicdoor's exact product behavior. See the automatic model selection pillar for the category and Auto versus manual selection for control tradeoffs.

The exact decision order

Jev returns probabilities for two typed yes/no questions. Magicdoor uses a 0.30 quality threshold and 0.50 execution threshold. These are evaluated policy thresholds, not universal difficulty scores or percentages of expected quality improvement.

Worked examples

Routing is fresh per turn, but effective saved history matters. An Opus answer can be followed by a simple DeepSeek transformation in a compatible text conversation; a prior non-image file or container context can block an Opus handoff even if the latest prompt is text-only. Selecting any model manually from the model lineup bypasses Jev.

What Jev sees

The routing state contains up to six recent user/assistant text messages, including the latest request, plus the latest request and a capability flag. It excludes file parts and file URLs, tool payloads, system messages, separate reasoning parts, and embedded <think> blocks.

Capability checks use effective saved history, even though file content is excluded from Jev's text payload. The serialized state has a 16,000 UTF-8-byte cap. Oversized state is not truncated to justify a cheaper route. User-written text can still contain sensitive strings or URLs, so this boundary is not full redaction.

Failure, cancellation, and billing

Jev routing usage is operational telemetry and is not added to user answer-model usage. If answer-provider recovery runs after a failed attempt, each metered concrete model attempt is recorded separately. Auto has no invented fifteenth-model rate and is not a public API model.

Evidence and limitations

Magicdoor tested 164 synthetic scenarios with 324 live TypeSafe requests. This validates routing policy behavior, not real answer-quality superiority. The study did not generate and compare final answers. Auto also has four possible answer models, bounded context may miss relevant facts, and fail-closed rules can select Sol or Opus 5.5 for a simple-looking request.

The Jev explainer separates Magicdoor's routing evidence from TypeSafe's vendor claims.

Compare the categories directly: OpenRouter Auto vs Magicdoor Auto, ChatGPT automatic selection vs Magicdoor Auto, the published router ranking methodology, and when a multi-model workspace beats separate subscriptions.

Try Experimental Auto in chat

The Jev data-flow boundary explains what text reaches the router; the original test design distinguishes label agreement from answer quality.

FAQ

Is Magicdoor Auto a separate AI model?

No. Auto is an Experimental chat setting. It chooses a concrete answer model for each message. Usage is recorded against each concrete answer-model attempt, including any separately metered provider-recovery attempt, never against Auto.

Does Magicdoor Auto reconsider the model on every message?

Yes. Each Auto message is routed fresh. The previous answer model does not lock the next turn to the same model.

What happens if Jev cannot return a valid decision in time?

Missing, invalid, or oversized context, service errors, and the 2.5-second deadline fail closed to GPT-6.1 Sol for incompatible effective context or Claude Opus 5.5 for compatible text. User cancellation stops the request without launching another model.

Sources

Accessed September 23, 2026.

• Current Magicdoor routing policy, router, model registry, tests, and production notes.
• Magicdoor's recorded discovery report for methodology and limitations.
• TypeSafe API reference and Noul documentation.