Gemini 3 vs Opus 4.5 vs GPT-5.1 vs Grok 4: same-prompt 3D diorama showdown
cedric_chee · x · 2026-10-17
Google's official account kicked off a "3D isometric cute diorama games" challenge, admitting Gemini 3 is "very hard to beat". Multiple accounts ran the same prompt through Gemini 3 Pro, Claude Opus 4.5, GPT-5.1 and Grok 4, asking each to build an interactive 3D voxel island in a single HTML file.
Results: Gemini 3 delivered a complete interactive experience in one shot with the most polished visuals; Grok 4.1 Fast surprisingly took best aesthetics with superior UI and seasonal effects; Opus 4.5 was fastest (55 seconds) but "pan-fried" the island into a pancake with roads and parking lots crowding out nature; GPT-5.1 kept crashing and needed heavy babysitting. After several rounds Gemini 3 Pro and Grok 4.1 Fast tied for first — and Grok 4.1 Fast costs roughly one-eighth of Gemini 3 Pro, making it a compelling budget pick.
More from coding & agent
- Internship diary: feeding meeting transcripts to Claude to act as his mentor — beglen · 2026-10-17
- Users hit Grok Bot's long-context limits: 'retire' each bot after one project — alexcovo_eth · 2026-10-17
- Backend dev lands first $6k/month client with sub-200ms voice agent for collections — retarded_raj · 2026-10-17
- Open-source NotebookLM alternative Notestack adds an MCP server for Claude — phicreative1997 · 2026-10-17
- Quant Level Beats Decode Speed: 12-Task Benchmark of Qwen Flash Next Quants and Temperature — Dreeew84 · 2026-10-17
- SDLC is becoming ADLC: Limestone's 7-step agentic development pipeline, human-gated — alex_verem · 2026-10-17