-
Ornith spans 9B to 397B and writes its own training tasks
From DeepReinforce, Ornith hands task writing, scaffolding and solution writing to the model itself, comes in three sizes from 9B to 397B, and ships under MIT with a 256K context and local deployment.
-
Pavo reads a production system first, then sends a team of agents to tune the feed and the ranking
A continuous improvement platform for production AI systems: it distils code, warehouse, dashboards and experiments into a knowledge book with citations, then sends a team of agents to investigate in parallel and prove changes against a benchmark, handing back work that can ship.
-
HandyBot answers the phone for tradespeople and writes the job straight into the calendar
An AI phone answering service built for on-site trades in the United States: calls are picked up around the clock, the job is qualified, the appointment goes into the owner calendar, and follow ups and call records come with it, at 199 dollars a month with a seven day trial and no card.
-
Muse: Meta hands an AI its own cloud computer so it can actually do the errands
Meta's personal agent Muse runs on a separate cloud machine with its own browser: it fills forms, compares prices and places orders, keeps working after you close the app, and asks Sentinel before anything irreversible.
-
Let an AI answer your calls, book the table and hold the card: Cue gives every agent its own identity
An agent app from Manus for everyday life: every agent gets its own mailbox, phone number, wallet and cloud computer, so it can take calls, pay within a budget you set, and work in groups.
-
Treat an AI like a longtime colleague: Hermes Agent writes its own skills and remembers your habits
An open source autonomous agent from Nous Research: finished tasks harden into reusable skills that keep improving, preferences live in local files that survive sessions, and it plugs into more than twenty chat platforms.
-
An AI worker that lives in your repository: one @ in CodeBuddy NPC and the pull request arrives
A cloud agent from Tencent Cloud: mention it in an issue or pull request and it reads the repository and CI context, plans, edits code, opens a pull request and gets the build green, on its own.
-
Chatterfly: hold Fn, speak, and get a finished draft
A Tencent voice input tool in beta since September 2026: hold Fn, say it the way you would say it out loud, and get a cleaned up written version that already matches the audience you are writing for.
-
ArtCraft: build the scene first, let the model render it
An AI image and video tool that swaps prompt roulette for stage building: layer compositing, three dimensional scenes, character poses, and more than sixty models in one canvas.
-
Octop: a self hosted AI assistant that keeps the data at home
An AI assistant platform from Tencent Cloud, MIT licensed, where conversations, workspaces and keys stay on your own machine and nine chat platforms route into the same set of experts.
-
DeepSeek Harness: an open agent runtime you can install and run
An open source agent runtime from DeepSeek with a desktop build that bundles Node.js, pnpm and a Python environment, four preset modes, a plugin page and a trace view.
-
Continue review: an AI coding assistant where you pick the model
An open extension that plugs into the editor you already use and talks to whichever model you point it at: cloud, local, or a private deployment inside the company network, with the configuration written in a file you own.
-
Devin Desktop: after the Windsurf rename, an AI editor built as an agent control room
Devin Desktop is Cognition's AI coding environment, renamed from Windsurf in 2026, where a board called Agent Command Center supervises local and cloud agents side by side.
-
Fotor puts photo editing, design and collage in one browser workbench
One workbench for retouching, design and collage, running in a browser and as a desktop program. One tap enhancement, portrait tools, template driven posters and social images, and batch adjustments. The basics are free; advanced features and watermark removal need a subscription.
-
Harpa AI review: it summarises video and answers mail, and bills by the million tokens
One shortcut calls several frontier models onto any page, and the engine underneath is what turns them into page-aware commands, mail drafts and scheduled automation. The meter runs on tokens, so the tier you pick decides how much work you can hand over.
-
第 1 / 12 页