AI assistants that say which AI model they use
Which AI assistants tell you what model is behind them, and does it show in their scores?
14 assistants ranked
Fifteen of the 47 assistants in our catalog say something about the model behind them; 32 say nothing. Fourteen of the fifteen took at least five of our ten tests. Muse leads at 9.0 (it names its own Muse Spark), then Neo at 8.1 (Claude, GPT and Gemini), Dots at 8.0 (GPT-6 Astra) and Ollie at 7.75 (Anthropic, Google and OpenAI). Gemini Spark names Gemini 4 Argon and averages 5.5, so a named model is not a guide to a good score.
Top picks
- 19.0MuseTop pick
Tested on 7 of 10Best: Purchasing10, Email10
- 28.1NeoBest free option
Tested on 10 of 10Best: Travel9, Phone calls9
- 38.0DotsFastest that's still good
Tested on 7 of 10Best: Email10, Online tasks9Time to finish 6m 51s
- 47.8
Tested on 8 of 10Best: Online tasks9, Routines9
- 57.2
Tested on 6 of 10Best: Travel9, Phone calls8
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 | Travel | Purchasing | Phone calls | Proactive | Routines | Connected apps | Permissions | Memory | Average | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Muse | 8 | 7 | 10 | — | 10 | — | 10 | 9 | 9 | — | 9.0 |
| 2 | Neo | 7 | 9 | 8 | 9 | 6 | 9 | 8 | 9 | 8 | 8 | 8.1 |
| 3 | Dots | 9 | 6 | 7 | N/A | 10 | 9 | — | — | 7 | 8 | 8.0 |
| 4 | Ollie | 9 | 7 | 7 | N/A | 7 | 8 | 9 | 7 | 8 | — | 7.8 |
| 5 | Luke | 6 | 9 | — | 8 | 7 | — | 7 | — | — | 6 | 7.2 |
| 6 | Bantr | 8 | 8 | — | 5 | — | 7 | 6 | 8 | 7 | 7 | 7.0 |
| 7 | Buoy | 7 | 6 | 8 | — | — | 4 | 7 | 6 | 7 | 5 | 6.3 |
| 8 | OpenInstinct | 10 | — | — | — | 3 | — | — | 6 | 7 | 4 | 6.0 |
| 9 | Friday | — | 5 | — | 5 | — | — | 6 | 7 | 6 | 6 | 5.8 |
| 10 | Gemini Spark | 6 | 8 | 6 | 6 | 8 | 4 | 4 | 5 | 4 | 4 | 5.5 |
| 11 | Del | 5 | 7 | 1 | — | — | — | — | 7 | 8 | 4 | 5.3 |
| 12 | The Shadow | — | 6 | 8 | — | 4 | — | 4 | 6 | 5 | 4 | 5.3 |
| 13 | OpenClaw | 7 | 3 | 6 | 3 | 4 | 5 | 7 | 7 | 4 | 5 | 5.1 |
| 14 | AVA | 3 | 5 | 6 | 7 | 3 | — | 3 | 7 | 4 | 4 | 4.7 |
How this report works
Included are assistants whose catalog entry (checked 2026-10-26) records a model or model provider. Buoy lists providers (Anthropic, OpenAI, Google Cloud AI, Perplexity) without naming a model; Airtap has fewer than five tests and is not ranked. The rest are ranked by the average of the tests they took, with at least five taken.
How to choose
- Check whether it names one model or lets you choose. Luke offers a choice of 200+ models, OpenClaw any hosted or local model, OpenInstinct any model through Vercel AI Gateway; Neo routes between Claude, GPT and Gemini.
- Read the wording. Bantr says Anthropic without naming a model, The Shadow says OpenAI, AVA says Google Gemini through Google Cloud Vertex AI, Del says open-source models hosted by US-based providers.
- Judge by the tests. Gemini Spark, a Google assistant that names its model, averages 5.5; OpenClaw, which lets you pick any model, averages 5.1 because the model is only part of the product.
Which AI assistants say what model they use?
Fifteen of 47: Muse, Ollie, OpenInstinct, OpenClaw, Buoy, The Shadow, Airtap, AVA, Bantr, Del, Dots, Friday, Gemini Spark, Luke and Neo.
Does the model decide how good an assistant is?
Not on our tests. Assistants that name their model range from 9.0 (Muse) to 4.7 (AVA); the scores measure finished jobs, not the model's size.
Which assistants let me pick the model?
Luke, OpenClaw and OpenInstinct; see the choose-the-model report.
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 sourceUse Ollie if you want a text-only assistant that quietly watches your inbox and calendar and pings you; use Ohai if you want an app-based shared household workspace the whole family uses together.
Open sourceFrequently asked questions
Which assistant is best for this?
Muse leads with an average of 9.0, ahead of Neo and Dots.
Is there a good free option?
Muse is the highest-ranked assistant with a free plan, averaging 9.0.
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.