AI coding agents · scenario rankings

Boundary probe: personal agent runtimes (not in the grid)

Most Popular Self-Hosted Personal AI Agents

This is a boundary probe, not a category ranking: it asks about self-hosted personal agent runtimes, which barely overlap with coding agents (entity crossover with the category question is 0.0% in Chinese and 2.9% in English). Chinese-side completeness is 76.7%, so read it with that discount.

unstable definition

Measured 2026-09-22 · 12-seat model panel · 10 scenarios · methodology #2 v2026-09-22.1

Download image (with QR code)Share on X Share on LinkedIn

WeChat: save the image and long-press the QR code in a chat.

The ten

Rank	Entity	Mention rate (of answers)	Wilson 95% interval	Seats that mentioned it	Seats with stable coverage
1	AutoGPT	72.4%	[65.3%, 78.5%]	12 of 12	11 of 12
2	Dify	71.8%	[64.7%, 78.0%]	12 of 12	9 of 12
3	AnythingLLM	51.1%	[43.8%, 58.5%]	12 of 12	6 of 12
4	Flowise	51.1%	[43.8%, 58.5%]	11 of 12	7 of 12
5	OpenWebUI	50.6%	[43.2%, 57.9%]	12 of 12	7 of 12
6	CrewAI	47.1%	[39.9%, 54.5%]	12 of 12	2 of 12
7	Ollama	39.1%	[32.1%, 46.5%]	11 of 12	4 of 12
8	Open Interpreter	37.9%	[31.1%, 45.3%]	11 of 12	3 of 12
9	LangChain	33.9%	[27.3%, 41.2%]	11 of 12	2 of 12
10	SuperAGI	33.9%	[27.3%, 41.2%]	11 of 12	3 of 12

How it is counted: mention rate = answers naming it ÷ 174 (5 phrasings × 3 repeats × 12 seats = 180 calls, 174 valid); the last two columns count seats, and "stable coverage" means more than half of that seat's 15 answers. Different denominators, so 12 of 12 next to a rate below 100% is not a contradiction.

The sample behind this cell

Valid samples / calls expected: 174 / 180
Completeness: 96.7%
Effective entities (effN): 28.0
Entities mentioned at least once: 110

How this cell was asked

This cell used 5 phrasings, each asked 3 times. Every call asked for the ten names only — no reasons, no scores.

Most Popular Self-Hosted Personal AI Agents
Self-hosted personal AI agents that live in a chat app — what are the options?
I want a personal AI agent running on my own machine that handles messages and everyday tasks. Which open-source projects should I look at?
self hosted personal ai agent open source
List some open-source personal AI agent runtimes that people self-host.

Where this page stops

This is one measurement, not a trend line. The next run of the same phrasings comes next quarter.
Each call collected ten names at most. Anything past the tenth name is invisible in this cell, which is not the same as being absent.
Mention rate is the share of samples in this cell where a name came up. It is not market share, and it is not a judgement about the product.
Read this cell for who is present, not for the order.

Methodology: the panel, the phrasings and how completeness is counted

You can also query this data from Claude, Cursor or another agent, with the same reading rules attached. For agents →