Arbiter is an experimental hosted coding route for Orrery subscribers. It is built for coding agents that need clearer evidence, usage receipts, and review boundaries before a result is trusted.
The workspace, selected route, and review requirement stay explicit before hosted work begins.
Request received · boundary visibleArbiter is a subscriber-only hosted route. Capacity, latency, quotas, and availability can change during the beta.
Route named · availability not assumedSummary, usage, proof, and boundary records remain distinct so a polished answer cannot masquerade as verification.
Evidence visible · success not fabricatedThe receipt gives you material to inspect. It does not decide whether the work is safe to keep, run, or ship.
Receipt presented · acceptance remains yoursArbiter is a hosted Orrery route for coding tasks where evidence matters. The experience is simple: ask Orrery to run work, review what happened, see the usage it consumed, and decide whether to trust the result.
Inside Orrery, Arbiter appears as a hosted model route named arbiter-0.1-beta. It can be selected from Nexus alongside DeepSeek API and Doubleword hosted routes. Subscribers see usage meters, run receipts, and proof/boundary summaries where the app supports them.
The route is designed for agentic coding workflows: planning, editing, follow-up fixes, and review. It does not replace your own testing, code review, security review, or backups.
Most chat models produce the next token directly from the visible prompt and context. Arbiter is positioned differently in Orrery: it is tested as a coding route that values structured run signals such as task state, review evidence, and verifier-style feedback, then turns those signals into ordinary agent output.
The practical difference for subscribers is simple: Arbiter is meant for coding work where a final answer is not enough. Nexus should show the prompt, route, run summary, usage, proof, and boundary receipts so the user can inspect the result instead of trusting a raw completion.
Current evidence is internal, limited, and not presented here as a public benchmark. Matched evaluations will be published only with the task set, method, verifier configuration, and limitations needed to interpret them. Until then, the beta makes no leaderboard claim and guarantees no individual result.
Arbiter access is available for Orrery Pro, Max, and Ultra subscriber testing through hosted credits when the beta route is online. Signed-out users, unsubscribed accounts, and canceled or unpaid accounts cannot use hosted Arbiter credits.
The exact capacity, latency, quotas, model mix, and availability may change during the beta.
Treat Arbiter output like any other AI-generated code: inspect it, run tests, review security implications, and keep backups. Orrery can help organize evidence, but you remain responsible for what you accept, run, and ship.