# Nova Agent OS lifts Opus 4.8 to 83.9% on Terminal-Bench 2.1

> Original title: Nova got Opus 4.8 to edge past Fable 5 on Terminal-Bench 2.1.

- Source: Modal
- Published: 2026-10-29T15:28:20.000Z
- Bad Influence: https://badinfluence.news/items/kwxbtpvz7g2mcx1uf3x7uxflf
- Original: https://x.com/modal/status/2115828087405388096

## Summary

Modal 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.

## Post

Nova got Opus 4.8 to edge past Fable 5 on Terminal-Bench 2.1. 

Modal Sandboxes let the team decouple Nova’s agent loop and automatic review from tool execution, with fast startup and changes deployed in seconds to keep iteration quick. 

The result: 83.9%, up five percentage points from Opus 4.8’s Claude Code baseline and just ahead of Fable 5’s 83.8%.

## Quoted post @CarlHasselskog

> We made Opus 4.8 perform like Fable 5.
> 
> Same model. No fine-tuning.
> 
> Terminal-Bench 2.1:
> Opus 4.8 in Claude Code: 78.9%
> Opus 4.8 in Nova Agent OS: 83.9%
> Fable 5: 83.8%
> Built together with our friends at @modal. The frontier isn't only the model anymore.
> 
> https://novaagentos.ai/news/how-nova-beat-fable-5-with-opus-4-8

https://x.com/CarlHasselskog/status/2115541152204259453