AITests & Research
Tests & Research
Every test we run, with prompts, settings and outputs published, failures included.
Tests & Research
Our Test Lab
Original tests with every output published, failures included. Read each full test:
Every Test
All published tests, newest first.
21 Oct 2026How Good Is AI at Jumping? A Jump-Rope Physics Test
5 Oct 2026Best AI Coding Harnesses 2026: TGK Top 10 and Tier List
29 Sep 2026Best AI Models in 2026: TGK's Top 10 and Full Tier List
28 Sep 2026Can a Non-Developer Build an App With Codex?
9 Sep 2026How to Compare AI Models: The TGK Testing Method
4 Sep 2026AI Fight Scene Test: How to Write a Better Prompt
3 Sep 2026How Much Does a Usable 10-Second AI Video Cost?
1 Sep 2026Best AI Video Generators in 2026: Tested and Ranked
25 Aug 2026Best AI Image Generators in 2026: Tested and Ranked
13 Aug 2026Can Grok 4.6 Beat Claude Opus 5 and GPT-5.6 Sol? We Tested It
12 Aug 2026Best AI Image-to-Video Generators 2026: 6 Models Tested
10 Aug 2026Seedance 2.5 Review: Brilliant for 5 Seconds, Sloppy at 30
28 Jul 2026Microsoft Mage-Flow Test: Fast Mode Was Fast. Quality Mode Never Worked
27 Jul 2026Nano Banana Pro Prompting Guide: How to Get Better Results Without Overcomplicating Your Prompts
26 Jul 2026How to Create a Consistent AI Character Without LoRA Training
24 Jul 2026We Tested 7 AI Image Editors Across 36 First-Try Edits
22 Jul 2026I’ve Used VPNs All Over the World. Here’s What Still WorksHow We Rank AI Models
The three kinds of evidence behind our AI models ranking. Each lab chapter above shows how that test was run.
