{"slug": "have-we-made-a-unicorn-continuous-svg-pelican-style-benchmark", "title": "Have we made a unicorn? Continuous SVG-pelican style benchmark", "summary": "A human judge has been testing AI models daily by asking them to draw a unicorn in pure SVG, and no model has succeeded in 10 days. The benchmark will continue until a machine produces a drawing indistinguishable from a human's.", "body_md": "Have we made a unicorn?\n\n# NO\n\nNo for 10 days and counting.\n\nEvery day, the latest AI models are asked to draw a unicorn in pure SVG. A human judges the results. The verdict, so far, is no — and it will stay that way until a machine draws one you can’t tell from a human’s.\n\n## Today’s attempts\n\n- Anthropic Claude Haiku Latest\n- Anthropic Claude Sonnet Latest\n- Claude Opus 4.6 (Fast)\n- Claude Opus 4.7\n- Claude Opus 4.7 (Fast)\n- Claude Opus 4.8\n- Claude Opus 4.8 (Fast)\n- Claude Opus Latest\n- DeepSeek V4 Flash\n- DeepSeek V4 Pro\n- DeepSeek V4 Pro\n- DeepSeek V4 Pro\n- DeepSeek V4 Pro\n- GLM 5.1\n- GPT Chat Latest\n- GPT-5.5\n- GPT-5.5 Pro\n- Gemini 3.1 Flash Lite\n- Gemini 3.5 Flash\n- Gemini 3.5 Flash\n- Gemini 3.5 Flash\n- Gemma 4 26B A4B\n- Gemma 4 31B (free)\n- Google Gemini Flash Latest\n- Google Gemini Pro Latest\n- Granite 4.1 8B\n- Grok 4.3\n- Grok Build 0.1\n- Grok Build 0.1\n- Grok Build 0.1\n- Hy3 preview\n- Kimi K2.6\n- Kimi K2.7 Code\n- LFM2.5-1.2B-Instruct (free)\n- Laguna M.1 (free)\n- Laguna XS.2 (free)\n- Ling-2.6-1T\n- Ling-2.6-flash\n- Llama 4 Maverick\n- MiMo-V2.5\n- MiniMax M3\n- Mistral Medium 3.5\n- Mistral Medium 3.5\n- Mistral Medium 3.5\n- Mistral Medium 3.5\n- MoonshotAI Kimi Latest\n- Nemotron 3 Nano 30B A3B (free)\n- Nemotron 3 Super (free)\n- Nemotron 3 Ultra\n- Nemotron 3 Ultra (free)\n- Nemotron Nano 12B 2 VL (free)\n- Nemotron Nano 9B V2 (free)\n- Nex-N2-Pro (free)\n- OpenAI GPT Latest\n- OpenAI GPT Mini Latest\n- Perceptron Mk1\n- Qwen3 Next 80B A3B Instruct (free)\n- Qwen3.5 Plus 2026-04-20\n- Qwen3.6 27B\n- Qwen3.6 35B A3B\n- Qwen3.6 Flash\n- Qwen3.6 Max Preview\n- Qwen3.7 Max\n- Qwen3.7 Plus\n- Qwen3.7 Plus\n- Qwen3.7 Plus\n- Qwen3.7 Plus\n- Ring-2.6-1T\n- Step 3.7 Flash\n- gpt-oss-120b (free)\n- gpt-oss-20b (free)", "url": "https://wpnews.pro/news/have-we-made-a-unicorn-continuous-svg-pelican-style-benchmark", "canonical_source": "https://havewemadeaunicorn.com/", "published_at": "2026-06-14 08:47:06+00:00", "updated_at": "2026-06-14 09:00:41.529943+00:00", "lang": "en", "topics": ["artificial-intelligence", "generative-ai", "ai-research"], "entities": ["Anthropic", "Claude", "DeepSeek", "OpenAI", "GPT", "Gemini", "Google", "Mistral"], "alternates": {"html": "https://wpnews.pro/news/have-we-made-a-unicorn-continuous-svg-pelican-style-benchmark", "markdown": "https://wpnews.pro/news/have-we-made-a-unicorn-continuous-svg-pelican-style-benchmark.md", "text": "https://wpnews.pro/news/have-we-made-a-unicorn-continuous-svg-pelican-style-benchmark.txt", "jsonld": "https://wpnews.pro/news/have-we-made-a-unicorn-continuous-svg-pelican-style-benchmark.jsonld"}}