A-Fantasia Backwards Spelling
General QA · 2025-04-27
A-Fantasia Backwards Spelling gives a dictionary definition and asks for the defined noun spelled backwards. The 100 words are sampled from constrained WordNet nouns using Brown-corpus frequency filters; scores are error rates, so lower is better.
Top models (lower is better)
| Model | Score |
|---|---|
| Opus 4.1 | 15.0 |
| Opus 4.5 | 16.0 |
| Opus 4.6 | 17.0 |
| Opus 4 | 22.0 |
| GPT-5.6 Sol | 23.0 |
| Sonnet 4.5 | 24.0 |
| GPT-5.4 | 26.0 |
| GPT-5.5 | 27.0 |
| Gemini 3 Pro Preview | 27.0 |
| Gemini 3 Flash Preview | 28.0 |
| Opus 4.8 | 28.0 |
| GPT-4.5 | 32.0 |
| Sonnet 4 | 32.0 |
| GPT-5.6 Terra | 33.0 |
| Claude 3.5 Sonnet (Oct 2024) | 34.0 |
| Claude 3.7 Sonnet | 35.0 |
| Opus 3 | 35.0 |
| Grok 3 | 36.0 |
| GPT-5.2 | 37.0 |
| Gemini 3.1 Pro Preview | 38.0 |
| Sonnet 4.6 | 39.0 |
| DeepSeek-V4-Pro | 45.0 |
| Qwen3.7-Max | 46.0 |
| Gemini 2.5 Pro | 51.0 |
| GPT-4o | 57.0 |
| GPT-5 Chat (2025-08-07) | 62.0 |
| Inkling | 62.0 |
| GPT-5.1 | 63.0 |
| GPT-4.1 | 64.0 |
| Gemini 2.5 Flash | 64.0 |
| Haiku 4.5 | 69.0 |
| Kimi K2.6 | 70.0 |
| GPT-5.6 Luna | 72.0 |
| Qwen3.6 Plus (2026-04-02) | 72.0 |
| Gemini 3.1 Flash-Lite | 74.0 |
| Qwen3.7-Plus | 74.0 |
| Gemini 2.0 Flash 001 | 77.0 |
| GLM-5.2 | 77.0 |
| GLM-5 | 78.0 |
| GLM-5.1 | 78.0 |