Skip to content
Trending storyDeveloping

Eigent AI introduces AgenticRL-Bench, a benchmark for agentic reinforcement learning

1 article1 sourcesince Nov 11Last article

Overview

AISummary of 1 article

Eigent AI, through The Environment Lab, has introduced AgenticRL-Bench, a benchmark of 22 tasks rebuilt from real reinforcement learning incidents in systems including Miles, Slime, Verl, SGLang and Harbor.

Per the report, on the 19 verifiable tasks none of the six agents passed more than 14.0% of attempts. On one open-ended task, the best agent raised pass@1 from 0 to 0.186 using a self-developed method on 4×H200 GPUs over 10 to 16 hours. RadixArk, which says Miles helped power the effort, congratulated the team. The Environment Lab's announcement describes AgenticRL-Bench as the first release from its effort at Eigent AI.

Written by AI from the articles below · updated Nov 11, 5:23 PM ET

Check the sources:

Latest developmentNov 11, 5:00 PM ET
Eigent AI introduces AgenticRL-Bench, a benchmark for agents doing agentic RL

Article timeline

The articles in this story. Times are ET.

Nov 11
  1. RadixArkOfficial
    Eigent AI introduces AgenticRL-Bench, a benchmark for agents doing agentic RL

    AIEigent AI, through The Environment Lab, introduced AgenticRL-Bench with 22 tasks rebuilt from real RL incidents in systems including Miles, Slime, Verl, SGLang and Harbor. On the 19 verifiable tasks, none of the six agents passed more than 14.0% of attempts. On one open-ended task, the best agent improved pass@1 from 0 to 0.186 using a self-developed method on 4×H200 GPUs over 10–16 hours. RadixArk, which says Miles helped power the effort, congratulated the team.

Heat Trend

Current comparable heat
7
Peak comparable heat
10Nov 11, 6:00 PM ET
Change in past 24 hours
–

The trend compares only continuously observed participants, so its scope may be smaller than the current heat total. Hover or tap for hourly heat; use arrow keys with a keyboard.