Skip to content
Trending storyDeveloping

Nova Agent OS scores 83.9% on Terminal-Bench 2.1 with Opus 4.8

1 article1 sourcesince Oct 29Last article 1h ago ·

Overview

AISummary of 1 article

Modal says Nova Agent OS lifted Opus 4.8 to 83.9% on Terminal-Bench 2.1, five points above the 78.9% Opus 4.8 scored in Claude Code and just above Fable 5's 83.8%.

Modal says Nova ran its agent loop and automatic review in Modal Sandboxes, separate from tool execution, and that the model received no fine-tuning. The result comes from Modal's own report and has not been independently verified.

Written by AI from the articles below · updated Oct 29, 11:33 AM ET

Check the sources:

Latest developmentOct 29, 11:28 AM ET
Nova Agent OS lifts Opus 4.8 to 83.9% on Terminal-Bench 2.1

Article timeline

The articles in this story. Times are ET.

Oct 29
  1. ModalOfficial
    Nova Agent OS lifts Opus 4.8 to 83.9% on Terminal-Bench 2.1

    AIModal says Nova used Modal Sandboxes to run its agent loop and automatic review separately from tool execution, lifting Opus 4.8 to 83.9% on Terminal-Bench 2.1. That is five percentage points above Opus 4.8's 78.9% in Claude Code and just ahead of Fable 5's 83.8%. Modal says the model received no fine-tuning.

Heat Trend

Not enough continuous observations to show a trend yet.