Published AI API benchmark evidence

AI Meeting Assistants

This page shows the edition’s published aggregate context. Approved excerpts are shown where attached; full raw provider responses, private prompts and internal review notes remain private.

Edition date: 12 October 2026
Geography / locale: UK / en-GB
Executions: 192 valid / 192 planned
Collection window: 12 October 2026 – 12 October 2026
Snapshot reference: 91c7e0d72692a1618b1614190593051f89b487fdadc6184b6be7cc5d12587c90

Observed AI recommendations under methodology public-benchmark-v1.1.0. This is not a product-quality rating.

Scope: 19 products scored, 15 held, 1 excluded for this edition. Fellow was moved from the scored list to an edition hold after collection, because no saved evidence captured before the 11 Oct 2026 22:59 UTC cutoff verified that it could be bought in the UK. This was decided on evidence alone, not on results; a private 20-product calculation existed before this change. Fellow's mentions do not score in this edition, every other product's observed answer positions are unchanged, and Fellow is not permanently ineligible.

Gemini API results in this edition were collected and analysed under a founder risk acceptance without written permission from Google; Google permission was requested and not received. Results in this edition are shown under neutral provider labels. One provider family was included under a founder risk acceptance: its terms restrict named public disclosure and written permission has not been received.

No publication-safe supporting excerpts are attached to this edition. The table below provides aggregate results; it does not provide answer-level evidence for independent inspection.

Published score table

Diagnostics explain observations but are not additional ranking components.

Display correction: engine coverage reflects valid data from all four providers for this edition. Published scores and ranks are unchanged.

Published AI Meeting Assistants benchmark rankings
Rank	Product	RankHigh Score	Recommended	First choice	Engine coverage	Confidence	Movement	Benchmark date	Evidence
1	Fireflies.ai	49.24	86.5%	26.6%	4 / 4	—	Not comparable	12 October 2026	—
2	Fathom	35.89	60.4%	21.9%	4 / 4	—	Not comparable	12 October 2026	—
3	tl;dv	19.61	53.6%	4.2%	4 / 4	Statistically close—not clearly separated	Not comparable	12 October 2026	—
4	Otter.ai	19.48	62.5%	1%	4 / 4	Statistically close—not clearly separated	Not comparable	12 October 2026	—
5	Microsoft Teams Copilot	12.66	18.8%	9.4%	4 / 4	—	Not comparable	12 October 2026	—
6	Zoom AI Companion	7.67	15.1%	2.6%	4 / 4	Statistically close—not clearly separated	Not comparable	12 October 2026	—
7	Granola	6.28	22.4%	0.5%	4 / 4	Statistically close—not clearly separated	Not comparable	12 October 2026	—
8	Read AI	3.78	11.5%	1%	4 / 4	Statistically close—not clearly separated	Not comparable	12 October 2026	—
9	Jamie	3.71	12.5%	0%	4 / 4	Statistically close—not clearly separated	Not comparable	12 October 2026	—
10	MeetGeek	3.65	7.3%	2.1%	4 / 4	Statistically close—not clearly separated	Not comparable	12 October 2026	—
11	Webex AI Assistant	2.86	4.7%	1.6%	4 / 4	Statistically close—not clearly separated	Not comparable	12 October 2026	—
12	HappyScribe	1.66	5.2%	0%	4 / 4	Statistically close—not clearly separated	Not comparable	12 October 2026	—
13	Gemini in Google Meet	1.39	4.2%	0%	4 / 4	Statistically close—not clearly separated	Not comparable	12 October 2026	—
14	Sembly AI	1.01	3.6%	0%	4 / 4	Statistically close—not clearly separated	Not comparable	12 October 2026	—
15	Notta	0.54	1.6%	0%	4 / 4	Statistically close—not clearly separated	Not comparable	12 October 2026	—
16	Krisp	0.51	2.1%	0%	4 / 4	Statistically close—not clearly separated	Not comparable	12 October 2026	—
T17	Circleback Tied result Not observed in the top five during this edition.	0.00	0%	0%	4 / 4	Statistically close—not clearly separated	Not comparable	12 October 2026	—
T17	Tactiq Tied result Not observed in the top five during this edition.	0.00	0%	0%	4 / 4	Statistically close—not clearly separated	Not comparable	12 October 2026	—
T17	Laxis Tied result Not observed in the top five during this edition.	0.00	0%	0%	4 / 4	Statistically close—not clearly separated	Not comparable	12 October 2026	—

1
Fireflies.ai
Published · 12 October 2026
Score
49.24
Recommended
86.5%
First choice
26.6%
Confidence
—
Not comparable
2
Fathom
Published · 12 October 2026
Score
35.89
Recommended
60.4%
First choice
21.9%
Confidence
—
Not comparable
3
tl;dv
Published · 12 October 2026
Score
19.61
Recommended
53.6%
First choice
4.2%
Confidence
Statistically close—not clearly separated
Not comparable
4
Otter.ai
Published · 12 October 2026
Score
19.48
Recommended
62.5%
First choice
1%
Confidence
Statistically close—not clearly separated
Not comparable
5
Microsoft Teams Copilot
Published · 12 October 2026
Score
12.66
Recommended
18.8%
First choice
9.4%
Confidence
—
Not comparable
6
Zoom AI Companion
Published · 12 October 2026
Score
7.67
Recommended
15.1%
First choice
2.6%
Confidence
Statistically close—not clearly separated
Not comparable
7
Granola
Published · 12 October 2026
Score
6.28
Recommended
22.4%
First choice
0.5%
Confidence
Statistically close—not clearly separated
Not comparable
8
Read AI
Published · 12 October 2026
Score
3.78
Recommended
11.5%
First choice
1%
Confidence
Statistically close—not clearly separated
Not comparable
9
Jamie
Published · 12 October 2026
Score
3.71
Recommended
12.5%
First choice
0%
Confidence
Statistically close—not clearly separated
Not comparable
10
MeetGeek
Published · 12 October 2026
Score
3.65
Recommended
7.3%
First choice
2.1%
Confidence
Statistically close—not clearly separated
Not comparable
11
Webex AI Assistant
Published · 12 October 2026
Score
2.86
Recommended
4.7%
First choice
1.6%
Confidence
Statistically close—not clearly separated
Not comparable
12
HappyScribe
Published · 12 October 2026
Score
1.66
Recommended
5.2%
First choice
0%
Confidence
Statistically close—not clearly separated
Not comparable
13
Gemini in Google Meet
Published · 12 October 2026
Score
1.39
Recommended
4.2%
First choice
0%
Confidence
Statistically close—not clearly separated
Not comparable
14
Sembly AI
Published · 12 October 2026
Score
1.01
Recommended
3.6%
First choice
0%
Confidence
Statistically close—not clearly separated
Not comparable
15
Notta
Published · 12 October 2026
Score
0.54
Recommended
1.6%
First choice
0%
Confidence
Statistically close—not clearly separated
Not comparable
16
Krisp
Published · 12 October 2026
Score
0.51
Recommended
2.1%
First choice
0%
Confidence
Statistically close—not clearly separated
Not comparable
T17
Circleback
Tied result
Not observed in the top five during this edition.
Published · 12 October 2026
Score
0.00
Recommended
0%
First choice
0%
Confidence
Statistically close—not clearly separated
Not comparable
T17
Tactiq
Tied result
Not observed in the top five during this edition.
Published · 12 October 2026
Score
0.00
Recommended
0%
First choice
0%
Confidence
Statistically close—not clearly separated
Not comparable
T17
Laxis
Tied result
Not observed in the top five during this edition.
Published · 12 October 2026
Score
0.00
Recommended
0%
First choice
0%
Confidence
Statistically close—not clearly separated
Not comparable

Read Methodology B Challenge this edition