MCP-Tasks (BenchLM): leaderboard
Interactive tasks using Model Context Protocol tools and integrations, as reported in Qwen launch comparisons and mirrored by BenchLM. The published task score is a distinct protocol from MCP Atlas. These are published results, not BenchLM ability estimates.
Metric: Task Score (%). Source: benchlm.ai. Saturation forecast: Around February 2027. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3.5 397B A17B | 74.2 | #154 |
| 2 | Qwen 3.6 Plus | 74.1 | #124 |
| 3 | Claude Opus 4.5 | 71.8 | #87 |
| 4 | GLM-5 | 60.8 | #149 |
| 5 | Kimi K2.5 | 59.1 | #152 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=mcp-tasks-benchlm · How It Works · Data refreshed daily, snapshot 2026-11-05.