What's new in 1.2.2: Streaming replies now put ~99% less load on Conol's servers, keeping things fast and steady even with many people chatting at once. Read more: conol.ai/blog/release-1.2.2

What's new in 1.2.1: Files you attach while answering the agent now reach it, attachments show as clean file cards, Billing spending history loads again. Read more: conol.ai/blog/release-1.2.1

What's new in 1.2.0: Spending history on the Billing page, credits used shown right in chat, Claude Haiku 5.5 + Gemini 4 Argon + Gemini 3.8 Flash + DeepSeek V4.1 Flash added, voice input fixed. Read more: conol.ai/blog/release-1.2.0

What's new in 1.1.3: Welcome credits now arrive once you verify your email, quest rewards unlock after your first day. Read more: conol.ai/blog/release-1.1.3

What's new in 1.1.2: SVG, HEIC and other image formats arrive as files. Read more: conol.ai/blog/release-1.1.2

What's new in 1.1.1: Chat attachments reach the agent, plus attachment cards in the chat. Read more: conol.ai/blog/release-1.1.1

What's new in 1.1.0: Version history for notes, undo any agent edit, recently deleted recovery, plus Claude Sonnet 5.5 and GPT-6.1 Sol. Read more: conol.ai/blog/release-1.1.0

What's new in 1.0.0: Conol reaches version 1.0! A calmer, single-line view of the agent at work, plus a Mac multi-account sign-in fix. Read more: conol.ai/blog/release-1.0.0

What's new in 0.59.0: GLM 5.3 FlashX joins the model menu — same 1M context and image reading as Flash, tuned for speed. Google sign-in on iPhone finally works. Read more: conol.ai/blog/release-0.59.0

What's new in 0.58.0: You can now see when the agent is waiting for you, plus 4 new models — GPT-6 Sol, GPT-6 Luna, Claude Opus 5.5 and Fable 5.1. Read more: conol.ai/blog/release-0.58.0

What's new in 0.57.0: DeepSeek V4 Flash Vision reads your images, several models cost less, and DeepSeek V4 Pro is back online. Read more: conol.ai/blog/release-0.57.0

What's new in 0.56.0: Notes and chats now live side by side in tabs, the Mac app opens as many windows as you want, and CoHDL code is syntax highlighted. Read more: conol.ai/blog/release-0.56.0

What's new in 0.55.0: GPT-6 Astra joins the model picker, agent edits to open notes no longer vanish, and you can cancel a timer from any chat. Read more: conol.ai/blog/release-0.55.0

What's new in 0.54.0: press ⌘K to go anywhere, keyboard shortcuts everywhere, and GLM 5.3 Flash joins the model picker at a fraction of the price. Read more: conol.ai/blog/release-0.54.0

What's new in 0.53.0: reorder your workspaces, GLM 5.3 joins the model picker, and agents now see workspaces created mid-conversation. Read more: conol.ai/blog/release-0.53.0

What's new in 0.52.0: attach files up to 200MB, jump straight back to the newest message, and failed turns now explain what went wrong. Read more: conol.ai/blog/release-0.52.0

What's new in 0.51.0: stay signed in to multiple accounts and switch without logging out, bring your agent into a Slack channel with @mentions, plus Qwen 3.8 Max joins the model list. Read more: conol.ai/blog/release-0.51.0

What's new in 0.50.0: connect Attio and let your agent work your CRM — look up contacts, log calls, update deals. Plus reasoning now streams live as it happens. Read more: conol.ai/blog/release-0.50.0

What's new in 0.49.0: Claude Opus 5 joins the model picker as the new flagship — a clear step up on hard, multi-step reasoning and coding at the same price. Plus an official Discord community. Read more: conol.ai/blog/release-0.49.0

What's new in 0.48.1: pin the chats you use most, give any chat your own title that sticks, and quick chats now name themselves. Read more: conol.ai/blog/release-0.48.1