Entidex · Signal Layer
We probe Claude, ChatGPT, Gemini, Grok and Perplexity at regular intervals and grade their answers against a known ground truth. The result is a live picture of LLM awareness lag — per model, per topic, with retrieval on or off.
Tracking awareness scores for ATP Finals over recent probes.
How well each LLM knows ATP Finals — by topic and retrieval mode.
| Model | Retrieval OFF | Retrieval ON |
|---|---|---|
| ●Claude | 85 | 79 |
| ●ChatGPT | 85 | 82 |
| ●Gemini | 85 | 85 |
| ●Grok | 85 | 82 |
| ●Perplexity | 85 | 85 |
| Model | Retrieval OFF | Retrieval ON |
|---|---|---|
| ●Claude | 69 | 79 |
| ●ChatGPT | 75 | 82 |
| ●Gemini | 72 |
| Model | Retrieval OFF | Retrieval ON |
|---|---|---|
| ●Claude | 69 | 69 |
| ●ChatGPT | 82 | 82 |
| ●Gemini | 82 | 85 |
| ○Grok | 65 | 32 |
| ●Perplexity | 82 | 85 |
| ●Grok | 69 | 79 |
| ●Perplexity | 82 | 85 |
