TODO for AI
vs 
Notion
By Marcell Havlik, founder of TODO for AI · Updated
The docs-and-databases workspace, with Custom Agents that run on schedules and triggers inside it. Here is where each one wins, with numbers, and who should pick which.
| Fact | Notion | TODO for AI |
|---|---|---|
| Interface | Web, desktop, mobile | Web, desktop, mobile, CLI, IDE (Zed, JetBrains, VS Code), voice |
| Models | Claude, GPT, Gemini, Grok, you pick | Any provider: Claude, GPT, Gemini, Grok, Ollama |
| Source | Closed | Edge + CLI public |
| Starts at | Free | Free |
Key takeaways
- Models. Notion model choice: Claude, GPT, Gemini, Grok, you pick. TODO for AI runs any provider, Claude, GPT, Gemini, Grok or a local Ollama model, on one plan, and one model can review another's work.
- Benchmark. TODO for AI scored 89.9% on Terminal-Bench 2.1 with Opus 5.5 (high) (own harness, 80/89 tasks, 2026-10-08; not an official leaderboard submission).
- Price. Notion pricing: Free, Plus about $10, Business $20 per member per month billed yearly ($24 monthly), Enterprise custom. Custom Agents run on Notion credits at $10 per 1,000. TODO for AI has a free plan (Claude Sonnet only), then Starter $20, Pro $100 or Ultra $200 a month with any provider included.
- Where the work happens. Notion gets one task done inside the workspace. TODO for AI is one board your team and agents work through together, on your own PC, a cloud VM and your logged-in Chrome.
- Who should pick which. Pick Notion if your company runs on Notion and the work is documents, databases and reporting. Pick TODO for AI if the work needs a shell, a repo, a deploy or a script.
Notion
Claude, GPT, Gemini, Grok, you pick
TODO for AI
Any providerWe are not competing for the same job
Notion gets the task done.
TODO for AI organises the whole company’s work.
Notion answers one question: how do I get this task done, inside the workspace? So do the others — Cursor inside the editor, Claude Code in one person's session, Lindy by asking its assistant.
We answer a different one: where does the work live? One board per project, and a project for each part of the company — engineering, marketing, ops, finance, one per client. Your colleagues work on it. So do your agents, as many as you want to run, each with its own model, tools and permissions. Items carry the people who worked them and the agent that ran them, and the board filters by person, so what the company has in flight is one screen instead of a Slack thread.
That is not a team plan bolted onto an AI tool. It is the product. The agents are how items get closed; the board is what the company runs on.
What Notion is, and where it stops
Notion 3.3 Custom Agents are much more than a writing assistant. They run on a schedule or a trigger, choose between Claude, GPT, Gemini and Grok, read from Notion and 18+ connected apps over MCP, optionally browse the web, hand work to each other, and keep an activity log. You share them like people, with permission levels. On the collaboration axis Notion is genuinely ahead of every agent vendor.
The limit is the perimeter. A Notion agent lives in the workspace and reaches out through connections an admin wired. It has no shell, no machine of its own, and no way into your logged-in browser. If the task is to update a database, file a bug, post a report or message a channel, it is excellent. If the task is to run the deploy, fix the script, or log into a vendor portal that has no MCP server, it cannot start.
Teams whose work already lives in Notion and who want agents to maintain it.
Three reasons Notion users look elsewhere
Nothing in Notion runs a command. The agent cannot open a terminal on your laptop, on a cloud VM or in a sandbox, so a task whose real form is a command line only happens if some connected service exposes it as an action.
Reach is the MCP servers an admin enabled plus optional web browsing. A tool without an MCP server, or a portal that needs your logged-in session, stays out of reach.
Agent work draws Notion credits at $10 per 1,000 on top of the per-member price, so the more the agents do the less predictable the bill.
Six dimensions that decide it
Notion 19/30TODO for AI 29/30How does Notion compare to TODO for AI, feature by feature?
| Feature | Notion | TODO for AI |
|---|---|---|
Persistent cloud VM An always-on machine. Files, logins and sessions survive between runs. | No | Yes |
Model choice | Claude, GPT, Gemini, Grok | Any provider |
Cross-vendor review Review GPT output with Claude, or the other way around. | No | Yes |
Drives your real browser Browser extension (Chrome, Firefox) plus cloud browsers. | Agent web browsing, not your session | Yes |
Business tasks, not only code SEO, email, ads, social, CRM via CLI tools. | Docs, databases, connected apps | Yes |
Any CLI as an integration | MCP only | Any binary |
Voice control | No | Yes |
People and agents on one board Colleagues and agents work the same list. Items show who contributed and which agent ran them; filter the board by person. | Yes, agents are shareable | Yes |
Many agents, each configured Each agent gets its own model, prompt, tools and permissions. Pick which one takes an item. | Custom Agents, per-agent model | Unlimited, per task |
Projects for the whole company Engineering, marketing, ops, finance: separate projects and agents, one team and one bill. | Teamspaces | Unlimited, shared |
Team roles, pooled usage | Per seat + credits | Roles + pooled |
Tool permissions | Share levels + tool access | Allow, approve, block |
Open runtime | No | Edge + CLI public |
Same ladder, different meter
Seats for the workspace, then Notion credits for agent work at $10 per 1,000. Agent runs and Workers both draw on the credit pool.
One pool priced in dollars of model cost, spendable on any vendor. Session and weekly windows pace it, and a 1.3-1.5x plan multiplier applies, so a $20 plan is about $15 of list-price API. Team seat prices go into one shared pool.