Inference

APEX-Agents

A professional-services agent benchmark covering long-horizon knowledge-work tasks.

  • In index
  • Agentic
ModelsModels with a result.10
Top resultBest result on this test.57.5%Grok 4.6
Top-3 spreadPoints from first to third.19.7 pts
YearYear of release.N/A

Result and price

020406080100$0.1$0.3$1$3$10$30RESULTOUTPUT PRICE PER 1M TOKENS · LOG SCALE

Results

10 results
#ModelResultThe score from the source.
01Grok 4.657.5%
02Apodex 1.138.5%
03Step 5 Preview37.8%
04Kimi K337.6%
05Hy4 preview37.1%
06dots3-note Preview30.8%
07Ling 3.0 Flash Fin29.2%
08Apodex 1.1 Mini27.7%
09Solar Pro 418.7%
10Solar Open 216.6%