Verdict
Stable Beluga 2 leads 5–1 across 6 shared benchmarks.
internlm-20b 1 · Stable Beluga 2 5 · higher isn't always better — check the caveats below.
Reported scores — protocols may differ. How we compare
Commonsense Reasoning: Stable Beluga 2 leads 2–0 across 2 shared commonsense reasoning benchmarks (largest gap: HellaSwag, 78.8 vs 70.8 — single evaluator).
Stable Beluga 2 leads 5–1 across 6 shared benchmarks. internlm-20b leads 1 benchmark and Stable Beluga 2 leads 5. Higher isn't always better — see the integrity caveats on each benchmark.
internlm-20b and Stable Beluga 2 have 6 benchmarks in common in our data — those are the only rows where a direct, apples-to-apples comparison is drawn.
Stable Beluga 2 leads commonsense reasoning 2–0. Stable Beluga 2 leads reasoning 2–0.