ChatGPT alternatives that do personal tasks for you
Which assistants can I message like ChatGPT and have them actually do the task?
18 assistants ranked
These are the assistants you can reach by iMessage, SMS, WhatsApp or Telegram and that took at least three of four task tests: booking a table, buying an item, replying to email and running a scheduled job. 18 qualify. Muse leads at 9.5 over four, Pally follows at 8.25 and Ollie at 8.0. ChatGPT is not in our catalog and has no score here, so this report does not say how any of them compare with it.
Top picks
- 19.5MuseTop pick
Tested on 4 of 4Best: Purchasing10, Email10
- 28.3
Tested on 4 of 4Best: Routines10, Purchasing9Time to finish 7m 1s
- 38.0
Tested on 4 of 4Best: Online tasks9, Routines9
- 47.5
Tested on 4 of 4Best: Email9, Routines9
- 57.3
Tested on 3 of 4Best: Online tasks8, Purchasing7
Top pick: highest average here. Best free option: highest among assistants you can use for free. Fastest that's still good: quickest median time to finish among those averaging 7 or more.
Full ranking
| # | Assistant | Online tasks | Purchasing | Routines | Average | |
|---|---|---|---|---|---|---|
| 1 | Muse | 8 | 10 | 10 | 10 | 9.5 |
| 2 | Pally | 7 | 9 | 7 | 10 | 8.3 |
| 3 | Ollie | 9 | 7 | 7 | 9 | 8.0 |
| 4 | Waldo | 6 | 6 | 9 | 9 | 7.5 |
| 5 | Caddy | 8 | 7 | 7 | — | 7.3 |
| 6 | Neo | 7 | 8 | 6 | 8 | 7.3 |
| 7 | Tomo | 7 | 6 | 8 | — | 7.0 |
| 8 | Poke | 4 | — | 9 | 8 | 7.0 |
| 9 | Luke | 6 | — | 7 | 7 | 6.7 |
| 10 | Wajo / Fo | 2 | 8 | 7 | 7 | 6.0 |
| 11 | OpenClaw | 7 | 6 | 4 | 7 | 6.0 |
| 12 | Vellum | 4 | 4 | — | 8 | 5.3 |
| 13 | The Shadow | — | 8 | 4 | 4 | 5.3 |
| 14 | Ask Ranger | 4 | 7 | 4 | 6 | 5.3 |
| 15 | Alfi by Text.ai | — | 4 | 6 | 5 | 5.0 |
| 16 | Persona | 5 | 4 | 5 | 5 | 4.8 |
| 17 | tinyNature | 4 | 4 | 5 | 5 | 4.5 |
| 18 | Lucas | 4 | 5 | 3 | 5 | 4.3 |
How this report works
Included are assistants whose catalog channels (checked 2026-10-26) list iMessage, SMS, WhatsApp or Telegram; app-only assistants are left out. They are ranked by the average of Online task, Purchasing, Email replies and Routines, with at least three taken. Travel booking and Phone calls are left to the ai-assistant-that-gets-things-done report, which covers every channel.
How to choose
- Check the channel first. Pally is on iMessage, SMS, WhatsApp and Telegram; Ollie and Caddy on iMessage and SMS; Muse on its app and WhatsApp; Waldo on iMessage only.
- Look at each task, not only the average. Poke scores 9 on email replies but 4 on online tasks, and Wajo / Fo scores 8 on purchasing but 2 on online tasks.
- Check the price. Pally lists a free plan, Ollie lists $25 a month, Muse $20 and Poke $19, as checked on 2026-10-26; many others list a free tier without a price.
Is there a ChatGPT alternative that does things for me?
Eighteen assistants you can message took at least three of our four task tests. Muse (9.5), Pally (8.25) and Ollie (8.0) average highest. We do not score ChatGPT, so we make no comparison with it.
Can I use these without installing an app?
Many can: they work through iMessage, SMS, WhatsApp or Telegram. The ai-assistant-with-no-app-to-install report lists the ones that need no app at all.
Why are travel and phone calls not counted?
They are in the ai-assistant-that-gets-things-done report. Here the four tests are the ones an everyday message request is most likely to hit: a booking, a purchase, an email reply and a recurring job.
What users say
The agent, which Meta launched on September 8, booked a Friday date night and made a packing schedule for my upcoming move. It also emailed a friend about a trip next month and made a planning document with places we want to visit.
Open sourceWhen I asked for a list of places to go for date night, it pulled up two options in my neighborhood that had been closed for years.
Open sourceFrequently asked questions
Which assistant is best for this?
Muse leads with an average of 9.5, ahead of Pally and Ollie.
Is there a good free option?
Muse is the highest-ranked assistant with a free plan, averaging 9.5.
How did you test these assistants?
Every assistant got the same published tests, word for word, scored from three checkpoints and the number of corrections it needed. Each test is on the Dimensions page.