1,340
Business, management and finance · rank 182 of 402
- Unit
- Arena rating, higher is better
- Range
- 1,314 to 1,366
- Sample
- 437 votes
- Configuration
- Qwen3 32B
- 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
Expert prompts · rank 112 of 359
- Unit
- Arena rating, higher is better
- Range
- 1,361 to 1,435
- Sample
- 236 votes
- Configuration
- Qwen3 32B
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,331
Instruction following · rank 192 of 409
- Unit
- Arena rating, higher is better
- Range
- 1,312 to 1,350
- Sample
- 858 votes
- Configuration
- Qwen3 32B
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,347
Overall · rank 199 of 409
- Unit
- Arena rating, higher is better
- Range
- 1,337 to 1,356
- Sample
- 3926 votes
- Configuration
- Qwen3 32B
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,308
Writing, literature and language · rank 208 of 408
- Unit
- Arena rating, higher is better
- Range
- 1,291 to 1,325
- Sample
- 1121 votes
- Configuration
- Qwen3 32B
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,304
Unsupported claims · rank 186 of 407
- Unit
- claims per product, lower is better
- Range
- 1,281 to 1,326
- Sample
- 616 votes
- Configuration
- Qwen3 32B
- Measured
- 25 Sep 2026
- Not shown
76.4%
Irrelevance detection · rank 68 of 109
- Unit
- % correct, higher is better
- Configuration
- qwen3-32b
- Measured
- 16 Dec 2025
- Not shown
- Not general refusal or safety behaviour.
82.4%
Irrelevance detection · rank 45 of 109
- Unit
- % correct, higher is better
- Configuration
- qwen3-32b
- Measured
- 16 Dec 2025
- Not shown
- Not general refusal or safety behaviour.
26.7%
Memory · rank 28 of 109
- Unit
- % correct, higher is better
- Configuration
- qwen3-32b
- Measured
- 16 Dec 2025
- Not shown
- Not long-term personal memory in a product: sessions are BFCL's scripted ones.
15.7%
Memory · rank 55 of 109
- Unit
- % correct, higher is better
- Configuration
- qwen3-32b
- Measured
- 16 Dec 2025
- Not shown
- Not long-term personal memory in a product: sessions are BFCL's scripted ones.
47.9%
Multi-turn tasks · rank 18 of 109
- Unit
- % correct, higher is better
- Configuration
- qwen3-32b
- Measured
- 16 Dec 2025
- Not shown
- Not open-ended agent work: the tools and tasks are BFCL's simulated APIs.
43.2%
Multi-turn tasks · rank 25 of 109
- Unit
- % correct, higher is better
- Configuration
- qwen3-32b
- Measured
- 16 Dec 2025
- Not shown
- Not open-ended agent work: the tools and tasks are BFCL's simulated APIs.
48.7%
Overall accuracy · rank 29 of 109
- Unit
- % correct, higher is better
- Configuration
- qwen3-32b
- 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.
46.8%
Overall accuracy · rank 33 of 109
- Unit
- % correct, higher is better
- Configuration
- qwen3-32b
- 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.
93.8%
Relevance detection · rank 8 of 109
- Unit
- % correct, higher is better
- Configuration
- qwen3-32b
- Measured
- 16 Dec 2025
- Not shown
- Not whether the call itself was right; only that one was attempted.
81.2%
Relevance detection · rank 41 of 109
- Unit
- % correct, higher is better
- Configuration
- qwen3-32b
- Measured
- 16 Dec 2025
- Not shown
- Not whether the call itself was right; only that one was attempted.
88.8%
Single-turn calls (curated) · rank 11 of 109
- Unit
- % correct, higher is better
- Configuration
- qwen3-32b
- 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.
90.3%
Single-turn calls (curated) · rank 3 of 109
- Unit
- % correct, higher is better
- Configuration
- qwen3-32b
- 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.
82.0%
Single-turn calls (user-contributed) · rank 3 of 109
- Unit
- % correct, higher is better
- Configuration
- qwen3-32b
- 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.
82.0%
Single-turn calls (user-contributed) · rank 3 of 109
- Unit
- % correct, higher is better
- Configuration
- qwen3-32b
- 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.
21.5%
Reliably publish-ready · rank 41 of 109
- Unit
- % of products, higher is better
- Configuration
- qwen3-32b
- Measured
- 16 Dec 2025
- Not shown
- Not your catalogue: invented products with generated images, so a model's result on your own feed can differ.
19.0%
Requested-length error · rank 163 of 370
- Unit
- % off the requested word count, lower is better
- Configuration
- qwen3-32b
- Measured
- 12 Sep 2025
- Not shown
- Not other format limits such as character counts or bullet counts.
74.0%
Requested-length error · rank 333 of 370
- Unit
- % off the requested word count, lower is better
- Configuration
- qwen3-32b
- Measured
- 13 Sep 2025
- Not shown
- Not other format limits such as character counts or bullet counts.
0.32
Style adherence · rank 289 of 370
- Unit
- score from 0 to 1, higher is better
- Configuration
- qwen3-32b
- Measured
- 12 Sep 2025
- Not shown
- Not brand voice on your own examples: UGI's prompts are private and lean towards creative writing.
0.31
Style adherence · rank 318 of 370
- Unit
- score from 0 to 1, higher is better
- Configuration
- qwen3-32b
- Measured
- 13 Sep 2025
- Not shown
- Not brand voice on your own examples: UGI's prompts are private and lean towards creative writing.
33.0
Writing score · rank 249 of 370
- Unit
- score out of 100, higher is better
- Configuration
- qwen3-32b
- Measured
- 12 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.
30.3
Writing score · rank 268 of 370
- Unit
- score out of 100, higher is better
- Configuration
- qwen3-32b
- Measured
- 13 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.
99.9%
Answer rate · rank 19 of 108
- Unit
- % of documents, higher is better
- Configuration
- Qwen3 32B
- 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.
5.9%
Hallucination rate · rank 24 of 108
- Unit
- % of summaries, lower is better
- Configuration
- Qwen3 32B
- 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.
