1,350
Business, management and finance · rank 183 of 402
- Unit
- Arena rating, higher is better
- Range
- 1,338 to 1,362
- Sample
- 2241 votes
- Configuration
- DeepSeek V3
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,349
Creative writing · rank 150 of 407
- Unit
- Arena rating, higher is better
- Range
- 1,338 to 1,359
- Sample
- 3623 votes
- Configuration
- DeepSeek V3
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,348
Expert prompts · rank 190 of 359
- Unit
- Arena rating, higher is better
- Range
- 1,331 to 1,365
- Sample
- 1236 votes
- Configuration
- DeepSeek V3
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,343
Instruction following · rank 192 of 409
- Unit
- Arena rating, higher is better
- Range
- 1,336 to 1,350
- Sample
- 8606 votes
- Configuration
- DeepSeek V3
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,358
Overall · rank 192 of 409
- Unit
- Arena rating, higher is better
- Range
- 1,354 to 1,363
- Sample
- 21770 votes
- Configuration
- DeepSeek V3
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,356
Writing, literature and language · rank 159 of 408
- Unit
- Arena rating, higher is better
- Range
- 1,348 to 1,364
- Sample
- 5913 votes
- Configuration
- DeepSeek V3
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
17.0%
Requested-length error · rank 140 of 370
- Unit
- % off the requested word count, lower is better
- Configuration
- DeepSeek V3
- Measured
- 21 Sep 2025
- Not shown
- Not other format limits such as character counts or bullet counts.
0.37
Style adherence · rank 84 of 370
- Unit
- score from 0 to 1, higher is better
- Configuration
- DeepSeek V3
- Measured
- 21 Sep 2025
- Not shown
- Not brand voice on your own examples: UGI's prompts are private and lean towards creative writing.
43.2
Writing score · rank 188 of 370
- Unit
- score out of 100, higher is better
- Configuration
- DeepSeek V3
- Measured
- 21 Sep 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.
97.5%
Answer rate · rank 85 of 108
- Unit
- % of documents, higher is better
- Configuration
- DeepSeek V3
- 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.
6.1%
Hallucination rate · rank 25 of 108
- Unit
- % of summaries, lower is better
- Configuration
- DeepSeek V3
- 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.
