Platform benchmark

deepseek-v3.2-tee

deepseek · GPQA-Diamond · seed 42 · 2026-08-21 08:57 UTC

ResultsGPQA-Diamond · n=96 · seed 42

78.1%

78.1% correct · 100.0% answered on GPQA-Diamond

Score per $

4.26

Run cost

$0.18

Latency p50

217.1 s

Latency p95

548.0 s

Dataset: GPQA-Diamond · question and response text withheld under dataset access terms.

Per-item results96 items · tap a row for result details

Question

deepseek-v3.2-teedeepseek