1,414
Business, management and finance · rank 105 of 402
- Unit
- Arena rating, higher is better
- Range
- 1,402 to 1,426
- Sample
- 2493 votes
- Configuration
- DeepSeek V3.1
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,417
Business, management and finance · rank 98 of 402
- Unit
- Arena rating, higher is better
- Range
- 1,404 to 1,431
- Sample
- 1930 votes
- Configuration
- DeepSeek V3.1
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,388
Creative writing · rank 87 of 407
- Unit
- Arena rating, higher is better
- Range
- 1,374 to 1,401
- Sample
- 1960 votes
- Configuration
- DeepSeek V3.1
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,402
Creative writing · rank 73 of 407
- Unit
- Arena rating, higher is better
- Range
- 1,387 to 1,416
- Sample
- 1606 votes
- Configuration
- DeepSeek V3.1
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,426
Expert prompts · rank 98 of 359
- Unit
- Arena rating, higher is better
- Range
- 1,404 to 1,448
- Sample
- 725 votes
- Configuration
- DeepSeek V3.1
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,432
Expert prompts · rank 86 of 359
- Unit
- Arena rating, higher is better
- Range
- 1,407 to 1,457
- Sample
- 528 votes
- Configuration
- DeepSeek V3.1
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,402
Instruction following · rank 107 of 409
- Unit
- Arena rating, higher is better
- Range
- 1,392 to 1,412
- Sample
- 3663 votes
- Configuration
- DeepSeek V3.1
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,416
Instruction following · rank 86 of 409
- Unit
- Arena rating, higher is better
- Range
- 1,405 to 1,427
- Sample
- 2819 votes
- Configuration
- DeepSeek V3.1
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,417
Overall · rank 111 of 409
- Unit
- Arena rating, higher is better
- Range
- 1,411 to 1,423
- Sample
- 14543 votes
- Configuration
- DeepSeek V3.1
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,416
Overall · rank 111 of 409
- Unit
- Arena rating, higher is better
- Range
- 1,409 to 1,423
- Sample
- 11408 votes
- Configuration
- DeepSeek V3.1
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,398
Writing, literature and language · rank 92 of 408
- Unit
- Arena rating, higher is better
- Range
- 1,388 to 1,408
- Sample
- 3397 votes
- Configuration
- DeepSeek V3.1
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,407
Writing, literature and language · rank 80 of 408
- Unit
- Arena rating, higher is better
- Range
- 1,396 to 1,419
- Sample
- 2705 votes
- Configuration
- DeepSeek V3.1
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
30.0%
Requested-length error · rank 253 of 370
- Unit
- % off the requested word count, lower is better
- Configuration
- deepseek-chat-v3.1
- Measured
- 2 Oct 2025
- Not shown
- Not other format limits such as character counts or bullet counts.
34.0%
Requested-length error · rank 269 of 370
- Unit
- % off the requested word count, lower is better
- Configuration
- DeepSeek V3.1
- Measured
- 2 Oct 2025
- Not shown
- Not other format limits such as character counts or bullet counts.
0.34
Style adherence · rank 222 of 370
- Unit
- score from 0 to 1, higher is better
- Configuration
- deepseek-chat-v3.1
- Measured
- 2 Oct 2025
- Not shown
- Not brand voice on your own examples: UGI's prompts are private and lean towards creative writing.
0.36
Style adherence · rank 127 of 370
- Unit
- score from 0 to 1, higher is better
- Configuration
- DeepSeek V3.1
- Measured
- 2 Oct 2025
- Not shown
- Not brand voice on your own examples: UGI's prompts are private and lean towards creative writing.
51.1
Writing score · rank 153 of 370
- Unit
- score out of 100, higher is better
- Configuration
- deepseek-chat-v3.1
- Measured
- 2 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.
51.9
Writing score · rank 147 of 370
- Unit
- score out of 100, higher is better
- Configuration
- DeepSeek V3.1
- Measured
- 2 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.
94.5%
Answer rate · rank 93 of 108
- Unit
- % of documents, higher is better
- Configuration
- DeepSeek V3.1
- 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.
5.5%
Hallucination rate · rank 18 of 108
- Unit
- % of summaries, lower is better
- Configuration
- DeepSeek V3.1
- 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.
