Gemini 3.1 Pro Plays 3D Tetris Surprisingly Well
PromptInjection_ · reddit · 2026-07-20
This post showcases Gemini 3.1 Pro performing "unexpectedly well" in a 3D Tetris test. The author had low expectations but found the results quite solid, especially given it was an older version of 3.1 Pro. The accompanying image links to a full test article titled "3D Tetris Doesn't Care About Your LLM Benchmark Scores," implying that this mini-game test evaluates a model's practical spatial reasoning and manipulation skills rather than just traditional leaderboard metrics.
More from Models
- Korean startup says its model scored 44 on AAII and matches DeepSeek V4 Pro — JungWooHa2 · 2026-07-21
- OpenAI’s GPT-6 is predicted to be far more efficient than Fable — bindureddy · 2026-07-21
- Moonshot spotlights Kimi K3 and its API platform — pstAsiatech · 2026-07-21
- Motif 3 Beta lands on Hugging Face as South Korea’s foundation-model race heats up — Secure_Smoke_4280 · 2026-07-21
- Sakana AI’s Fugu-Cyber update tops real-world security benchmarks — SakanaAILabs · 2026-07-21
- Mythos release drama is being compared to o1, with limited rollout and an open-source clone — nptacek · 2026-07-21