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 Primer over recent probes.
How well each LLM knows Primer — by topic and retrieval mode.
| Model | Retrieval OFF | Retrieval ON |
|---|---|---|
| ·Claude | — | — |
| ●ChatGPT | 65 | — |
| ·Gemini | — | — |
| ·Grok | — | — |
| ●Perplexity | 85 | 85 |
| Model | Retrieval OFF | Retrieval ON |
|---|---|---|
| ·Claude | — | — |
| ●ChatGPT | 45 | — |
| ·Gemini | — | — |
| Model | Retrieval OFF | Retrieval ON |
|---|---|---|
| ·Claude | — | — |
| ●ChatGPT | 45 | — |
| ·Gemini | — | — |
| ·Grok | — | — |
| ●Perplexity | 69 | 85 |
| ·Grok | — | — |
| ●Perplexity | 79 | 85 |
