DRACO, 100 curated tasks; single model response per task; five Opus 4.6 judging runs per response, averaged.
条件
Adaptive thinking/max effort; web search/fetch, programmatic tool calling and code execution; 1M token limit, context compaction triggered at 200K. Judge change from original benchmark limits direct comparison to paper headline values.
Claude Opus 4.8 · DRACO · 100-task Perplexity deep-research benchmark · Scores and configurations
Claude Opus 4.8 record for DRACO · 100-task Perplexity deep-research benchmark. 1 score record for this release, including 1 exact value (may include historical records). Includes 1 source location.
DRACO, 100 curated tasks; single model response per task; five Opus 4.6 judging runs per response, averaged.
Conditions
Adaptive thinking/max effort; web search/fetch, programmatic tool calling and code execution; 1M token limit, context compaction triggered at 200K. Judge change from original benchmark limits direct comparison to paper headline values.