Benchmark Almanac

Stable Beluga 2

Stable Beluga 2.

Stability AI Released 20 Jul 2023
8
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
Stability AI Jul 2023

Stable Beluga 2

Open weights

117.0 #210 of 271
Epoch Capabilities Index
  • LAMBADA 71.3% #20 /26

Where it ranks

Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.

Knowledge and language

BenchmarkScoreRankSource
BoolQ 89.4% #5 of 75 Epoch AI
ARC-Challenge 86.1% #10 of 75 Epoch AI
HellaSwag 84.1% #13 of 74 Epoch AI
PIQA 83.3% #12 of 58 Epoch AI
BIG-Bench Hard 69.3% #12 of 45 Epoch AI
MMLU 68.6% #65 of 127 Epoch AI
LAMBADA 71.3% #20 of 26 Epoch AI

Math and science

BenchmarkScoreRankSource
GSM8K 69.6% #24 of 82 Epoch AI

Stability AI's models over time

Stability AI's models in color, against every other model on the Epoch Capabilities Index, by release date.

50 75 100 125 150 175 2023 2024 2025 2026 2027 Stable Beluga 2 50 75 100 125 150 175 2023 2024 2025 2026 2027 Stable Beluga 2