允许工具 / medium / 框架未记录 · DRACO; medium effort; web search/fetch, programmatic tools, code execution; 980k-token budget; own agent harness; five judge grading runs per response, mean; Claude Opus 4.6 grader.
DRACO · 100-task Perplexity deep-research benchmark · Normalized score · 允许工具 / medium / 框架未记录 · DRACO; medium effort; web search/fetch, programmatic tools, code execution; 980k-token budget; own agent harness; five judge grading runs per response, mean; Claude Opus 4.6 grader.
允许工具 / high / 框架未记录 · DRACO; high effort; web search/fetch, programmatic tools, code execution; 980k-token budget; own agent harness; five judge grading runs per response, mean; Claude Opus 4.6 grader.
DRACO · 100-task Perplexity deep-research benchmark · Normalized score · 允许工具 / high / 框架未记录 · DRACO; high effort; web search/fetch, programmatic tools, code execution; 980k-token budget; own agent harness; five judge grading runs per response, mean; Claude Opus 4.6 grader.
允许工具 / max / 框架未记录 · DRACO; max effort; web search/fetch, programmatic tools, code execution; 980k-token budget; own agent harness; five judge grading runs per response, mean; Claude Opus 4.6 grader.
DRACO · 100-task Perplexity deep-research benchmark · Normalized score · 允许工具 / max / 框架未记录 · DRACO; max effort; web search/fetch, programmatic tools, code execution; 980k-token budget; own agent harness; five judge grading runs per response, mean; Claude Opus 4.6 grader.
Regular display scope. Rankings are grouped by metric and evaluation protocol. Recorded tool access, budgets, and frameworks are shown separately. Scores remain available for reference when conditions are unspecified.
Normalized score %
Tools allowed / low / Framework not recorded · DRACO 100-task set; low effort; web search/fetch, programmatic tool calling, code execution; 980k-token budget.
Tools allowed / medium / Framework not recorded · DRACO 100-task set; medium effort; web search/fetch, programmatic tool calling, code execution; 980k-token budget.
DRACO · 100-task Perplexity deep-research benchmark · Normalized score · Tools allowed / medium / Framework not recorded · DRACO 100-task set; medium effort; web search/fetch, programmatic tool calling, code execution; 980k-token budget.
Tools allowed / high / Framework not recorded · DRACO 100-task set; high effort; web search/fetch, programmatic tool calling, code execution; 980k-token budget.
DRACO · 100-task Perplexity deep-research benchmark · Normalized score · Tools allowed / high / Framework not recorded · DRACO 100-task set; high effort; web search/fetch, programmatic tool calling, code execution; 980k-token budget.
Tools allowed / max / Framework not recorded · DRACO 100-task set; max effort; web search/fetch, programmatic tool calling, code execution; 980k-token budget.
DRACO · 100-task Perplexity deep-research benchmark · Normalized score · Tools allowed / max / Framework not recorded · DRACO 100-task set; max effort; web search/fetch, programmatic tool calling, code execution; 980k-token budget.
Tools allowed / medium / Framework not recorded · DRACO; 100 curated deep-research tasks; web search, web fetch, programmatic tool calling, code execution; 980k-token budget; five independent grader runs per response, averaged.
DRACO · 100-task Perplexity deep-research benchmark · Normalized score · Tools allowed / medium / Framework not recorded · DRACO; 100 curated deep-research tasks; web search, web fetch, programmatic tool calling, code execution; 980k-token budget; five independent grader runs per response, averaged.
Tools allowed / high / Framework not recorded · DRACO; 100 curated deep-research tasks; web search, web fetch, programmatic tool calling, code execution; 980k-token budget; five independent grader runs per response, averaged.
DRACO · 100-task Perplexity deep-research benchmark · Normalized score · Tools allowed / high / Framework not recorded · DRACO; 100 curated deep-research tasks; web search, web fetch, programmatic tool calling, code execution; 980k-token budget; five independent grader runs per response, averaged.
Tools allowed / max / Framework not recorded · DRACO; 100 curated deep-research tasks; web search, web fetch, programmatic tool calling, code execution; 980k-token budget; five independent grader runs per response, averaged.
DRACO · 100-task Perplexity deep-research benchmark · Normalized score · Tools allowed / max / Framework not recorded · DRACO; 100 curated deep-research tasks; web search, web fetch, programmatic tool calling, code execution; 980k-token budget; five independent grader runs per response, averaged.
Tools allowed / low / Framework not recorded · DRACO; low effort; web search/fetch, programmatic tools, code execution; 980k-token budget; own agent harness; five judge grading runs per response, mean; Claude Opus 4.6 grader.
DRACO · 100-task Perplexity deep-research benchmark · Normalized score · Tools allowed / low / Framework not recorded · DRACO; low effort; web search/fetch, programmatic tools, code execution; 980k-token budget; own agent harness; five judge grading runs per response, mean; Claude Opus 4.6 grader.
Tools allowed / medium / Framework not recorded · DRACO; medium effort; web search/fetch, programmatic tools, code execution; 980k-token budget; own agent harness; five judge grading runs per response, mean; Claude Opus 4.6 grader.
DRACO · 100-task Perplexity deep-research benchmark · Normalized score · Tools allowed / medium / Framework not recorded · DRACO; medium effort; web search/fetch, programmatic tools, code execution; 980k-token budget; own agent harness; five judge grading runs per response, mean; Claude Opus 4.6 grader.
Tools allowed / high / Framework not recorded · DRACO; high effort; web search/fetch, programmatic tools, code execution; 980k-token budget; own agent harness; five judge grading runs per response, mean; Claude Opus 4.6 grader.
DRACO · 100-task Perplexity deep-research benchmark · Normalized score · Tools allowed / high / Framework not recorded · DRACO; high effort; web search/fetch, programmatic tools, code execution; 980k-token budget; own agent harness; five judge grading runs per response, mean; Claude Opus 4.6 grader.
Tools allowed / xhigh / Framework not recorded · DRACO; xhigh effort; web search/fetch, programmatic tools, code execution; 980k-token budget; own agent harness; five judge grading runs per response, mean; Claude Opus 4.6 grader.
DRACO · 100-task Perplexity deep-research benchmark · Normalized score · Tools allowed / xhigh / Framework not recorded · DRACO; xhigh effort; web search/fetch, programmatic tools, code execution; 980k-token budget; own agent harness; five judge grading runs per response, mean; Claude Opus 4.6 grader.
Tools allowed / max / Framework not recorded · DRACO; max effort; web search/fetch, programmatic tools, code execution; 980k-token budget; own agent harness; five judge grading runs per response, mean; Claude Opus 4.6 grader.
DRACO · 100-task Perplexity deep-research benchmark · Normalized score · Tools allowed / max / Framework not recorded · DRACO; max effort; web search/fetch, programmatic tools, code execution; 980k-token budget; own agent harness; five judge grading runs per response, mean; Claude Opus 4.6 grader.