0.01
Confirmed task success · rank 15 of 46
- Unit
- IPS effect estimate, higher is better
- Range
- -0.02 to 0.04
- Sample
- 21948 observations
- Configuration
- Claude Sonnet 5
- Measured
- 28 Sep 2026
- Not shown
- Not a success rate: the estimated effect of choosing this model as the agent's orchestrator, against an even mix of the models tested, inside Arena's own harness.
0.09
Praise over complaint · rank 3 of 46
- Unit
- IPS effect estimate, higher is better
- Range
- 0.04 to 0.15
- Sample
- 6780 observations
- Configuration
- Claude Sonnet 5
- Measured
- 28 Sep 2026
- Not shown
- Not a success rate: the estimated effect of choosing this model as the agent's orchestrator, against an even mix of the models tested, inside Arena's own harness.
0.07
Steerability · rank 1 of 46
- Unit
- IPS effect estimate, higher is better
- Range
- 0.04 to 0.11
- Sample
- 22508 observations
- Configuration
- Claude Sonnet 5
- Measured
- 28 Sep 2026
- Not shown
- Not a success rate: the estimated effect of choosing this model as the agent's orchestrator, against an even mix of the models tested, inside Arena's own harness.
0.00
Tool grounding · rank 24 of 46
- Unit
- IPS effect estimate, higher is better
- Range
- 0.00 to 0.00
- Sample
- 3169153 observations
- Configuration
- Claude Sonnet 5
- Measured
- 28 Sep 2026
- Not shown
- Not a success rate: the estimated effect of choosing this model as the agent's orchestrator, against an even mix of the models tested, inside Arena's own harness.
1,476
Overall · rank 6 of 44
- Unit
- Arena rating, higher is better
- Range
- 1,468 to 1,483
- Sample
- 7411 votes
- Configuration
- Claude Sonnet 5 (high)
- Measured
- 13 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,196
Overall · rank 6 of 34
- Unit
- Arena rating, higher is better
- Range
- 1,189 to 1,203
- Sample
- 40230 votes
- Configuration
- Claude Sonnet 5
- Measured
- 24 Aug 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,452
Overall · rank 46 of 177
- Unit
- Arena rating, higher is better
- Range
- 1,448 to 1,455
- Sample
- 42969 votes
- Configuration
- Claude Sonnet 5 (high)
- 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.
1,465
Business, management and finance · rank 22 of 402
- Unit
- Arena rating, higher is better
- Range
- 1,458 to 1,472
- Sample
- 8542 votes
- Configuration
- Claude Sonnet 5 (high)
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,435
Creative writing · rank 35 of 407
- Unit
- Arena rating, higher is better
- Range
- 1,427 to 1,442
- Sample
- 8600 votes
- Configuration
- Claude Sonnet 5 (high)
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,512
Expert prompts · rank 9 of 359
- Unit
- Arena rating, higher is better
- Range
- 1,503 to 1,521
- Sample
- 4910 votes
- Configuration
- Claude Sonnet 5 (high)
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,467
Instruction following · rank 18 of 409
- Unit
- Arena rating, higher is better
- Range
- 1,461 to 1,473
- Sample
- 15428 votes
- Configuration
- Claude Sonnet 5 (high)
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,462
Overall · rank 44 of 409
- Unit
- Arena rating, higher is better
- Range
- 1,457 to 1,466
- Sample
- 43340 votes
- Configuration
- Claude Sonnet 5 (high)
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,452
Writing, literature and language · rank 27 of 408
- Unit
- Arena rating, higher is better
- Range
- 1,446 to 1,459
- Sample
- 11604 votes
- Configuration
- Claude Sonnet 5 (high)
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
3.0%
Requested-length error · rank 18 of 370
- Unit
- % off the requested word count, lower is better
- Configuration
- Claude Sonnet 5 (high)
- Measured
- 30 Jun 2026
- Not shown
- Not other format limits such as character counts or bullet counts.
3.0%
Requested-length error · rank 18 of 370
- Unit
- % off the requested word count, lower is better
- Configuration
- claude-sonnet-5
- Measured
- 30 Jun 2026
- Not shown
- Not other format limits such as character counts or bullet counts.
6.0%
Requested-length error · rank 41 of 370
- Unit
- % off the requested word count, lower is better
- Configuration
- claude-sonnet-5
- Measured
- 30 Jun 2026
- Not shown
- Not other format limits such as character counts or bullet counts.
3.0%
Requested-length error · rank 18 of 370
- Unit
- % off the requested word count, lower is better
- Configuration
- claude-sonnet-5
- Measured
- 30 Jun 2026
- Not shown
- Not other format limits such as character counts or bullet counts.
7.0%
Requested-length error · rank 48 of 370
- Unit
- % off the requested word count, lower is better
- Configuration
- claude-sonnet-5
- Measured
- 30 Jun 2026
- Not shown
- Not other format limits such as character counts or bullet counts.
0.37
Style adherence · rank 90 of 370
- Unit
- score from 0 to 1, higher is better
- Configuration
- Claude Sonnet 5 (high)
- Measured
- 30 Jun 2026
- Not shown
- Not brand voice on your own examples: UGI's prompts are private and lean towards creative writing.
0.35
Style adherence · rank 158 of 370
- Unit
- score from 0 to 1, higher is better
- Configuration
- claude-sonnet-5
- Measured
- 30 Jun 2026
- Not shown
- Not brand voice on your own examples: UGI's prompts are private and lean towards creative writing.
0.37
Style adherence · rank 104 of 370
- Unit
- score from 0 to 1, higher is better
- Configuration
- claude-sonnet-5
- Measured
- 30 Jun 2026
- Not shown
- Not brand voice on your own examples: UGI's prompts are private and lean towards creative writing.
0.38
Style adherence · rank 78 of 370
- Unit
- score from 0 to 1, higher is better
- Configuration
- claude-sonnet-5
- Measured
- 30 Jun 2026
- Not shown
- Not brand voice on your own examples: UGI's prompts are private and lean towards creative writing.
0.38
Style adherence · rank 67 of 370
- Unit
- score from 0 to 1, higher is better
- Configuration
- claude-sonnet-5
- Measured
- 30 Jun 2026
- Not shown
- Not brand voice on your own examples: UGI's prompts are private and lean towards creative writing.
56.8
Writing score · rank 121 of 370
- Unit
- score out of 100, higher is better
- Configuration
- Claude Sonnet 5 (high)
- Measured
- 30 Jun 2026
- Not shown
- Not business copy quality: UGI's prompts are private and lean towards creative writing, and models that often refuse get no score.
53.0
Writing score · rank 141 of 370
- Unit
- score out of 100, higher is better
- Configuration
- claude-sonnet-5
- Measured
- 30 Jun 2026
- Not shown
- Not business copy quality: UGI's prompts are private and lean towards creative writing, and models that often refuse get no score.
62.7
Writing score · rank 85 of 370
- Unit
- score out of 100, higher is better
- Configuration
- claude-sonnet-5
- Measured
- 30 Jun 2026
- Not shown
- Not business copy quality: UGI's prompts are private and lean towards creative writing, and models that often refuse get no score.
52.4
Writing score · rank 144 of 370
- Unit
- score out of 100, higher is better
- Configuration
- claude-sonnet-5
- Measured
- 30 Jun 2026
- Not shown
- Not business copy quality: UGI's prompts are private and lean towards creative writing, and models that often refuse get no score.
58.5
Writing score · rank 110 of 370
- Unit
- score out of 100, higher is better
- Configuration
- claude-sonnet-5
- Measured
- 30 Jun 2026
- Not shown
- Not business copy quality: UGI's prompts are private and lean towards creative writing, and models that often refuse get no score.