# MCP-Atlas mean claim coverage (MCP-Atlas claim coverage)

> Average coverage of required claims in answers produced during real-world MCP tool-use workflows.

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

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

## About MCP-Atlas claim coverage

- Year: 2026
- Tasks: Production-like multi-server MCP workflows
- Format: Mean claim coverage
- Difficulty: Real-world tool use
- Paper: [Claude Opus 5 System Card](https://www-cdn.anthropic.com/c5fbac3f0b1280a933ebd26d3cb8bb9f5bdeaf48/Claude%20Opus%205%20System%20Card.pdf)

Section 8.13.2 reports claim coverage alongside the separate MCP-Atlas pass rate and notes that production effort settings may have changed slightly since evaluation.

MCP-Atlas claim coverage is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Leaderboard (2 models)

| Rank | Model | Creator | Score |
|------|-------|---------|-------|
| 1 | [Claude Opus 5](/models/claude-opus-5) | Anthropic | 89.1% |
| 2 | [Pokee-Isaac 28B](/models/pokee-isaac-28b) | Pokee AI | 74.6% |

## FAQ

### What does MCP-Atlas claim coverage measure?

Average coverage of required claims in answers produced during real-world MCP tool-use workflows.

### Which model scores highest on MCP-Atlas claim coverage?

Claude Opus 5 by Anthropic currently leads with a score of 89.1% on MCP-Atlas claim coverage.

### How many models are evaluated on MCP-Atlas claim coverage?

2 AI models have been evaluated on MCP-Atlas claim coverage on BenchLM.

### Does MCP-Atlas claim coverage affect BenchLM's overall score?

Not directly. MCP-Atlas claim coverage is still displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Compare Top Models on MCP-Atlas claim coverage

- [Claude Opus 5 vs Pokee-Isaac 28B](/compare/claude-opus-5-vs-pokee-isaac-28b)