CompareOpen in App Store
★★★★★
Average rating on the App Store. Apple doesn't publish how many ratings each star has.
This app isn't in any of our charts yet.
Relay 1.5 updates the native Battle experience to match Relay’s current model catalog. • Adds Claude as an answer model and default independent cross-check. • Updates Gemini Judge attribution to Gemini 3.7. • Keeps ChatGPT, DeepSeek, Gemini, Grok, and Claude selectable where available. • Checks question access before starting a model request and keeps rejected drafts in the composer. • Includes reliability and model-attribution improvements.
Relay turns a question into a living knowledge thread you can compare, evaluate, and revisit. ASK • Choose two different AI models and a Judge from Relay's currently available model catalog. • Watch both independent answers stream into one Battle. • See the Judge's Winner or Tie, Why, what each model missed, and what both missed. • Continue naturally with follow-up questions. • Reopen completed private Battles from My Questions. DISCOVER • Browse public questions and complete two-answer Battle snapshots. • Read model attribution, Judge context, sources, and human perspectives. • Continue a public question privately in Ask. • Sign in to post a selected Battle snapshot or contribute a structured human evaluation. MY QUESTIONS • Keep private questions and notes local by default. • Sign in directly from My Questions for optional private cloud sync and My Arena history. • Sign out while keeping local notes, or delete the cloud copy. Publishing is always explicit. Public posts contain only the selected question, answers, and Judge context. Questions are processed by Relay's server and the selected AI providers. Time-sensitive questions may use web research. AI answers and Judge results can be wrong; verify important information.