1,363
Overall · rank 158 of 177
- Unit
- Arena rating, higher is better
- Range
- 1,357 to 1,370
- Sample
- 7656 votes
- Configuration
- Command A
- Measured
- 25 Sep 2026
- Not shown
- Not an error rate: claims that cannot be checked on the web are skipped, and preference still carries most of the weight.
1,351
Business, management and finance · rank 188 of 402
- Unit
- Arena rating, higher is better
- Range
- 1,345 to 1,358
- Sample
- 8770 votes
- Configuration
- Command A
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,336
Creative writing · rank 167 of 407
- Unit
- Arena rating, higher is better
- Range
- 1,328 to 1,343
- Sample
- 7549 votes
- Configuration
- Command A
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,344
Expert prompts · rank 197 of 359
- Unit
- Arena rating, higher is better
- Range
- 1,332 to 1,355
- Sample
- 2859 votes
- Configuration
- Command A
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,342
Instruction following · rank 194 of 409
- Unit
- Arena rating, higher is better
- Range
- 1,336 to 1,347
- Sample
- 15483 votes
- Configuration
- Command A
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,354
Overall · rank 199 of 409
- Unit
- Arena rating, higher is better
- Range
- 1,350 to 1,357
- Sample
- 55477 votes
- Configuration
- Command A
- 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
Writing, literature and language · rank 173 of 408
- Unit
- Arena rating, higher is better
- Range
- 1,342 to 1,354
- Sample
- 12723 votes
- Configuration
- Command A
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
84.2%
Irrelevance detection · rank 37 of 109
- Unit
- % correct, higher is better
- Configuration
- command-a
- Measured
- 16 Dec 2025
- Not shown
- Not general refusal or safety behaviour.
16.6%
Memory · rank 54 of 109
- Unit
- % correct, higher is better
- Configuration
- command-a
- Measured
- 16 Dec 2025
- Not shown
- Not long-term personal memory in a product: sessions are BFCL's scripted ones.
29.5%
Multi-turn tasks · rank 41 of 109
- Unit
- % correct, higher is better
- Configuration
- command-a
- Measured
- 16 Dec 2025
- Not shown
- Not open-ended agent work: the tools and tasks are BFCL's simulated APIs.
46.5%
Overall accuracy · rank 35 of 109
- Unit
- % correct, higher is better
- Configuration
- command-a
- Measured
- 16 Dec 2025
- Not shown
- Not a neutral average: the weighting is BFCL's. Not reliability on your own tools: BFCL's functions and queries are a fixed test set, and a correct call is judged by its form, not by what it achieved.
81.2%
Relevance detection · rank 41 of 109
- Unit
- % correct, higher is better
- Configuration
- command-a
- Measured
- 16 Dec 2025
- Not shown
- Not whether the call itself was right; only that one was attempted.
87.6%
Single-turn calls (curated) · rank 25 of 109
- Unit
- % correct, higher is better
- Configuration
- command-a
- Measured
- 16 Dec 2025
- Not shown
- Not reliability on your own tools: BFCL's functions and queries are a fixed test set, and a correct call is judged by its form, not by what it achieved.
78.5%
Single-turn calls (user-contributed) · rank 23 of 109
- Unit
- % correct, higher is better
- Configuration
- command-a
- Measured
- 16 Dec 2025
- Not shown
- Not reliability on your own tools: BFCL's functions and queries are a fixed test set, and a correct call is judged by its form, not by what it achieved.
46.5%
Web search · rank 28 of 109
- Unit
- % correct, higher is better
- Configuration
- command-a
- Measured
- 16 Dec 2025
- Not shown
- Not general research quality: questions have short, checkable answers.
21.0%
Requested-length error · rank 195 of 370
- Unit
- % off the requested word count, lower is better
- Configuration
- Command A
- Measured
- 26 Oct 2025
- Not shown
- Not other format limits such as character counts or bullet counts.
0.35
Style adherence · rank 196 of 370
- Unit
- score from 0 to 1, higher is better
- Configuration
- Command A
- Measured
- 26 Oct 2025
- Not shown
- Not brand voice on your own examples: UGI's prompts are private and lean towards creative writing.
40.6
Writing score · rank 212 of 370
- Unit
- score out of 100, higher is better
- Configuration
- Command A
- Measured
- 26 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.
97.6%
Answer rate · rank 84 of 108
- Unit
- % of documents, higher is better
- Configuration
- Command A
- 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.
9.3%
Hallucination rate · rank 47 of 108
- Unit
- % of summaries, lower is better
- Configuration
- Command A
- 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.
