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 Owala over recent probes.
How well each LLM knows Owala — by topic and retrieval mode.
| Model | Retrieval OFF | Retrieval ON |
|---|---|---|
| ●Claude | 79 | 79 |
| ●ChatGPT | 85 | 85 |
| ●Gemini | 82 | 85 |
| ●Grok | 76 | 79 |
| ●Perplexity | 85 | 85 |
| Model | Retrieval OFF | Retrieval ON |
|---|---|---|
| ●Claude | 69 | 79 |
| ●ChatGPT | 72 | 82 |
| ●Gemini | 72 |
| Model | Retrieval OFF | Retrieval ON |
|---|---|---|
| ●Claude | 66 | 69 |
| ●ChatGPT | 79 | 82 |
| ●Gemini | 82 |
| ●Grok | 79 | 79 |
| ●Perplexity | 62 | 82 |
| ●Grok | 82 | 82 |
| ●Perplexity | 82 | 75 |
