# Toolathlon Verified average assistant turns (Toolathlon Verified avg. turns)

> Average assistant turns per Toolathlon Verified trajectory.

Canonical page: https://benchlm.ai/benchmarks/toolathlonverifiedavgturns

- Category: [Agentic](/agentic)
- Last updated: September 27, 2026

## About Toolathlon Verified avg. turns

- Year: 2026
- Tasks: 108 verified real-world tool-use tasks
- Format: Average trajectory length
- Difficulty: Long-horizon application tool use
- Paper: [Claude Opus 5 System Card](https://www-cdn.anthropic.com/c5fbac3f0b1280a933ebd26d3cb8bb9f5bdeaf48/Claude%20Opus%205%20System%20Card.pdf)

Table 8.13.6.A reports trajectory length as context for the benchmark's long-horizon tool-use scores.

Toolathlon Verified avg. turns is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Leaderboard (3 models)

| Rank | Model | Creator | Score |
|------|-------|---------|-------|
| 1 | [Claude Opus 5.5](/models/claude-opus-5-5) | Anthropic | 26.9 turns |
| 2 | [Claude Fable 5.1](/models/claude-fable-5-1) | Anthropic | 23.7 turns |
| 3 | [Claude Opus 5](/models/claude-opus-5) | Anthropic | 23.5 turns |

## FAQ

### What does Toolathlon Verified avg. turns measure?

Average assistant turns per Toolathlon Verified trajectory.

### Which model scores highest on Toolathlon Verified avg. turns?

Claude Opus 5.5 by Anthropic currently leads with a score of 26.9 turns on Toolathlon Verified avg. turns.

### How many models are evaluated on Toolathlon Verified avg. turns?

3 AI models have been evaluated on Toolathlon Verified avg. turns on BenchLM.

### Does Toolathlon Verified avg. turns affect BenchLM's overall score?

Not directly. Toolathlon Verified avg. turns is still displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Compare Top Models on Toolathlon Verified avg. turns

- [Claude Opus 5.5 vs Claude Fable 5.1](/compare/claude-fable-5-1-vs-claude-opus-5-5)
- [Claude Fable 5.1 vs Claude Opus 5](/compare/claude-fable-5-1-vs-claude-opus-5)
