16.0%
Requested-length error · rank 127 of 370
- Unit
- % off the requested word count, lower is better
- Configuration
- grok-3
- Measured
- 8 Oct 2025
- Not shown
- Not other format limits such as character counts or bullet counts.
0.36
Style adherence · rank 127 of 370
- Unit
- score from 0 to 1, higher is better
- Configuration
- grok-3
- Measured
- 8 Oct 2025
- Not shown
- Not brand voice on your own examples: UGI's prompts are private and lean towards creative writing.
58.7
Writing score · rank 108 of 370
- Unit
- score out of 100, higher is better
- Configuration
- grok-3
- Measured
- 8 Oct 2025
- Not shown
- Not business copy quality: UGI's prompts are private and lean towards creative writing, and models that often refuse get no score.
93.0%
Answer rate · rank 97 of 108
- Unit
- % of documents, higher is better
- Configuration
- grok-3
- Measured
- 22 Sep 2026
- Not shown
- Not a quality score: a low rate usually means content filters were triggered, and hallucination rates are measured on answered documents only.
5.8%
Hallucination rate · rank 23 of 108
- Unit
- % of summaries, lower is better
- Configuration
- grok-3
- Measured
- 22 Sep 2026
- Not shown
- Not errors in open questions or other tasks: only summarisation, judged by Vectara's own model (HHEM-2.3), not by people, on news-style documents rather than your data.
