Models / granite-3.1-2b-instruct

IBM

granite-3.1-2b-instruct

6 published results from 1 source. Each card shows where the number comes from and what it does not measure. The overall leaderboard combines them; here each stands alone.

Provider
IBM
Sources
1
Our benchmarks
0
Price
Not yet published

Reported by others

1,213
Business, management and finance · rank 306 of 402
Unit
Arena rating, higher is better
Range
1,184 to 1,243
Sample
382 votes
Configuration
granite-3.1-2b-instruct
Measured
25 Sep 2026
Not shown
Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,147
Creative writing · rank 328 of 407
Unit
Arena rating, higher is better
Range
1,118 to 1,176
Sample
442 votes
Configuration
granite-3.1-2b-instruct
Measured
25 Sep 2026
Not shown
Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,218
Expert prompts · rank 280 of 359
Unit
Arena rating, higher is better
Range
1,183 to 1,254
Sample
225 votes
Configuration
granite-3.1-2b-instruct
Measured
25 Sep 2026
Not shown
Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,173
Instruction following · rank 332 of 409
Unit
Arena rating, higher is better
Range
1,157 to 1,190
Sample
1252 votes
Configuration
granite-3.1-2b-instruct
Measured
25 Sep 2026
Not shown
Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,179
Overall · rank 343 of 409
Unit
Arena rating, higher is better
Range
1,168 to 1,190
Sample
3188 votes
Configuration
granite-3.1-2b-instruct
Measured
25 Sep 2026
Not shown
Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
1,128
Writing, literature and language · rank 352 of 408
Unit
Arena rating, higher is better
Range
1,107 to 1,149
Sample
850 votes
Configuration
granite-3.1-2b-instruct
Measured
25 Sep 2026
Not shown
Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.