Blog

L Announcement 30 Jun 2026 · 1 min read Live translation and transcription get their own pages Speak, and be understood Two real-time speech tools now have their own dedicated pages, straight from the menu. Live Translation Pick two languages, start talking, and ToRun translates your speech in near real time — the… T Announcement 29 Jun 2026 · 1 min read The newest image, video and music models More choice, always current We keep the model catalogue current so you always have the newest options. Recent additions span images, video and music, alongside everything you already had. What's new - Images — the latest… V Product 26 Jun 2026 · 1 min read Voice Across Providers: Realtime, TTS, and Transcription Voice on ToRun is not one feature — it is three, and each one picks from multiple providers. Realtime conversations Live voice mode connects you to speech-native models for real back-and-forth conversation — interruption… B Product 23 Jun 2026 · 2 min read Bring Your Own Keys: Three Tiers, Real Savings If you already pay OpenAI, Anthropic, or Google directly, you should not pay full retail again to use those models through ToRun. Bring Your Own Key BYOK is how that works. How it works Add a provider API key in Settings… W Product 18 Jun 2026 · 2 min read Workflow Templates: 250+ Ready Starting Points A blank canvas is the most expensive thing we can show you. The workflow template library fixes the cold start: more than 250 workflows, each one a working automation you can run as-is or fork and edit. What is in the li… B Perspective 13 Jun 2026 · 2 min read Budgets Should Say No Politely Every AI platform has limits. The difference is what happens when you reach one. The common failure mode: a request quietly dies, an error code appears, and you are left guessing whether the model is down, your account i… D Product 9 Jun 2026 · 1 min read Deep Research: Answers With Receipts A chat answer is only as good as what you can verify. Deep Research mode exists for the questions where "sounds right" is not good enough. What it does differently Instead of answering from model memory, Deep Research: 1… P Product 4 Jun 2026 · 2 min read Projects Give Your Chats a Shared Memory Chats are great until the third one about the same topic. You paste the same background, re-explain the same constraints, and the assistant answers as if it just met you. Projects fix that. Create a project, drop your ch… T Perspective 30 May 2026 · 6 min read The Honest-UX Bar Every AI Product Should Clear Most AI products fail on a dimension that has nothing to do with model quality. They fail on honesty — not in the sense of lying outright, but in the quieter ways: burying what a call costs, expressing confident nonsense…