Questions · Models
The most overrated AI
7 of 10
models named GPT-4 · each model counted once, by its most frequent named answer
ChatGPT, Gemini and Grok got one model each. 3 models (Gemini 3.7 Flash and Qwen3.8 Max among them) gave the same answer in every reply.
Every model’s answer (10)
| Model | Answer | Replies |
|---|---|---|
| Claude Opus 5 | GPT-4 | 3 of 6 |
| DeepSeek V4 Pro 0423 | GPT-4 | 4 of 6 |
| GPT-5.6 Terra | GPT-4 | 6 of 6 · every reply |
| Gemini 3.7 Flash | GPT-4 | 5 of 5 · every reply |
| Grok 4.6 | GPT-4 | 3 of 6 |
| Large 2512 | GPT-4 | 3 of 5 |
| Qwen3.8 Max | GPT-4 | 6 of 6 · every reply |
| Llama 3.3 70B | ChatGPT | 3 of 4 |
| Kimi K3 | Gemini | 3 of 5 |
| GLM-5.3 | Grok | 2 of 3 |
How to read this
- 10 models gave a usable, named answer; 52 replies in all under default sampling. Refusals, errors and ties are not counted.
- “Every reply” marks a model that gave the same answer in every one of at least three replies.
- Part of a published study: read it.