News for hackers. Stuff that matters.

A comment on Show HN: TetrisBench – Gemini Flash reaches 66% win rate on Tetris against Opus.

Re:Show HN: TetrisBench – Gemini Flash reaches 66% win rate ... by vunderba on Monday January 26, 2026 @03:53PM (#46771367)
Interesting but frustratingly vague on details. How exactly are the models playing? Is it using some kind of PGN equivalent in Tetris that represents a on-going game, passing an ASCII representation, encoding as a JSON structure, or just directly sending screenshots of the game to the various LLMs?
[ Parent | on HN ]
Re:Show HN: TetrisBench – Gemini Flash reaches 66% win rate ... by storystarling on Monday January 26, 2026 @04:18PM (#46771687)
It has to be turn-based. Even with Flash's speed, the inference latency would kill you in a real-time loop. They're likely pausing the game state after every tick to wait for the API response before resuming.
[ Parent | on HN ]
Re:Show HN: TetrisBench – Gemini Flash reaches 66% win rate ... by ykhli on Monday January 26, 2026 @04:42PM (#46771967)
answered this in a comment above! It's not turn or visual layout based since LLMs are not trained that way. The representation is a JSON structure, but LLMs plug in algorithms and keeps optimizing it as the game state evolves
[ Parent | on HN ]
Re:Show HN: TetrisBench – Gemini Flash reaches 66% win rate ... by vunderba on Monday January 26, 2026 @05:37PM (#46772614)
Thanks for the clarification! Kind of reminds me of the Brian Moore's AI clocks which uses several LLMs to continuously generate HTML/CSS to create an analog clock for comparisons.

https://clocks.brianmoore.com

[ Parent | on HN ]
Re:Show HN: TetrisBench – Gemini Flash reaches 66% win rate ... by ykhli on Monday January 26, 2026 @06:17PM (#46773101)
Wow this is incredible!!
[ Parent | on HN ]
Re:Show HN: TetrisBench – Gemini Flash reaches 66% win rate ... by mhh__ on Monday January 26, 2026 @06:22PM (#46773153)
I suppose you could argue about whether it's an LLM at that point but vision is a huge part of frontier models now, no?
[ Parent | on HN ]
Re:Show HN: TetrisBench – Gemini Flash reaches 66% win rate ... by storystarling on Tuesday January 27, 2026 @04:21AM (#46777500)
Curious how the token economics compare here to a standard agent loop. It seems like if you're using the LLM as a JIT to optimize the algorithm as the game evolves, the context accumulation would get expensive fast even with Flash pricing.
[ Parent | on HN ]