Verdict
GPT-3.5 Turbo (Nov 2023) leads 4–2 across 6 shared benchmarks.
GPT-3.5 Turbo (Nov 2023) 4 · phi-3-mini 3.8B 2 · higher isn't always better — check the caveats below.
Reported scores — protocols may differ. How we compare
Knowledge: GPT-3.5 Turbo (Nov 2023) leads 2–0 across 2 shared knowledge benchmarks (largest gap: TriviaQA, 85.8 vs 64 — single evaluator).
Reasoning: Evenly split 1–1 across 2 shared reasoning benchmarks (largest gap: ARC AI2, 83.2 vs 79.87 — single evaluator).
GPT-3.5 Turbo (Nov 2023) leads 4–2 across 6 shared benchmarks. GPT-3.5 Turbo (Nov 2023) leads 4 benchmarks and phi-3-mini 3.8B leads 2. Higher isn't always better — see the integrity caveats on each benchmark.
GPT-3.5 Turbo (Nov 2023) and phi-3-mini 3.8B have 6 benchmarks in common in our data — those are the only rows where a direct, apples-to-apples comparison is drawn.
GPT-3.5 Turbo (Nov 2023) leads knowledge 2–0.