How TetrisBench Evaluates AI
stuffyokodraws · x · 2026-07-18
This post discusses the design and gameplay of TetrisBench: it provides an evaluation environment where models play Tetris, and users can directly compete against the models on the page.
Replies mention that when testing Fable against Kimi K3, Fable performed unexpectedly well in long-horizon planning and optimization; the video playback speed is 1.6x. The main focus here isn't on a specific model topping the leaderboard, but rather on how this benchmark translates capabilities like "planning and long-term optimization" into observable gameplay performance.
Related event: Fable Shows Impressive Planning Skills in TetrisBench(2 posts)→
More from Models
- Kimi K2.8 Preview rolls out: near-K3 coding performance, 1M context for all tiers — teortaxesTex · 2026-09-11
- Looking for a classifier of software engineering task shapes to pick models per task — StewartalsopIII · 2026-09-11
- DeepSeek V4 Pro API to continue after Sept 2026, billing unchanged — teortaxesTex · 2026-09-11
- DeepSeek V4.1 Flash Hits 98% of GPT-6 Astra's Score at 1.4% of the Cost in Third-Party Benchmark — ayushtweetshere · 2026-09-11
- TheZvi Polls: Has Your Coding Model Choice Changed Since Fable 5.1 and Astra? — TheZvi · 2026-09-11
- antirez Weighs In on Anthropic Banning Minors From Using Claude — antirez · 2026-09-11