1,386
Business, management and finance · rank 110 of 402
- Unit
- Arena rating, higher is better
- Range
- 1,356 to 1,416
- Sample
- 364 votes
- Configuration
- Gemma 3 12B
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,332
Creative writing · rank 154 of 407
- Unit
- Arena rating, higher is better
- Range
- 1,309 to 1,356
- Sample
- 600 votes
- Configuration
- Gemma 3 12B
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,274
Expert prompts · rank 225 of 359
- Unit
- Arena rating, higher is better
- Range
- 1,233 to 1,315
- Sample
- 186 votes
- Configuration
- Gemma 3 12B
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,321
Instruction following · rank 202 of 409
- Unit
- Arena rating, higher is better
- Range
- 1,305 to 1,337
- Sample
- 1145 votes
- Configuration
- Gemma 3 12B
- 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
Overall · rank 204 of 409
- Unit
- Arena rating, higher is better
- Range
- 1,332 to 1,351
- Sample
- 3829 votes
- Configuration
- Gemma 3 12B
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,313
Writing, literature and language · rank 203 of 408
- Unit
- Arena rating, higher is better
- Range
- 1,295 to 1,331
- Sample
- 1010 votes
- Configuration
- Gemma 3 12B
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
70.3%
Irrelevance detection · rank 83 of 109
- Unit
- % correct, higher is better
- Configuration
- gemma-3-12b-it
- Measured
- 16 Dec 2025
- Not shown
- Not general refusal or safety behaviour.
27.5%
Memory · rank 25 of 109
- Unit
- % correct, higher is better
- Configuration
- gemma-3-12b-it
- Measured
- 16 Dec 2025
- Not shown
- Not long-term personal memory in a product: sessions are BFCL's scripted ones.
5.8%
Multi-turn tasks · rank 79 of 109
- Unit
- % correct, higher is better
- Configuration
- gemma-3-12b-it
- Measured
- 16 Dec 2025
- Not shown
- Not open-ended agent work: the tools and tasks are BFCL's simulated APIs.
30.4%
Overall accuracy · rank 66 of 109
- Unit
- % correct, higher is better
- Configuration
- gemma-3-12b-it
- 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
- gemma-3-12b-it
- Measured
- 16 Dec 2025
- Not shown
- Not whether the call itself was right; only that one was attempted.
79.4%
Single-turn calls (curated) · rank 77 of 109
- Unit
- % correct, higher is better
- Configuration
- gemma-3-12b-it
- 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.
74.2%
Single-turn calls (user-contributed) · rank 46 of 109
- Unit
- % correct, higher is better
- Configuration
- gemma-3-12b-it
- 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.
4.0%
Web search · rank 67 of 109
- Unit
- % correct, higher is better
- Configuration
- gemma-3-12b-it
- Measured
- 16 Dec 2025
- Not shown
- Not general research quality: questions have short, checkable answers.
9.0%
Requested-length error · rank 67 of 370
- Unit
- % off the requested word count, lower is better
- Configuration
- Gemma 3 12B
- Measured
- 6 Sep 2025
- Not shown
- Not other format limits such as character counts or bullet counts.
0.29
Style adherence · rank 343 of 370
- Unit
- score from 0 to 1, higher is better
- Configuration
- Gemma 3 12B
- Measured
- 6 Sep 2025
- Not shown
- Not brand voice on your own examples: UGI's prompts are private and lean towards creative writing.
29.9
Writing score · rank 274 of 370
- Unit
- score out of 100, higher is better
- Configuration
- Gemma 3 12B
- Measured
- 6 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.
97.4%
Answer rate · rank 86 of 108
- Unit
- % of documents, higher is better
- Configuration
- Gemma 3 12B
- 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.
4.4%
Hallucination rate · rank 7 of 108
- Unit
- % of summaries, lower is better
- Configuration
- Gemma 3 12B
- 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.
