The story’s score and claims

Comments on No. 39 of 109 · AI News Report · edition of Friday 6 November 2026

GDPx agent benchmark: cost per point for 12 models, and the safety-filter caveat

Summary

SemiAnalysis released GDPx, a benchmark of 206 long, expert-level agent tasks across 11 fields, on November 2, 2026. The best model, Claude Opus 5.5, scores 12 percent at about $51 per task, while cheaper models like DeepSeek 4.1 Flash score far lower but cost under 20 cents per point.

Comments

Ranked by a public rubric, not by votes: what a comment adds counts, not who wrote it. How comments are ranked · Comment rules