agents benchmark
MCP-Mark Leaderboard
MCP-Mark evaluates LLMs on their ability to use Model Context Protocol (MCP) tools effectively across diverse MCP server scenarios.
Updated Sep 5, 2026
Models8
Model coverage8
MetricScore
EvidenceB
MCP-Mark Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelMA | Score81.10% | Percentile100.00% | Participants8 | EvidenceC | Evaluated |
| Rank02 | ModelAC | Score60.80% | Percentile85.71% | Participants8 | EvidenceC | Evaluated |
| Rank03 | ModelAC | Score58.70% | Percentile71.43% | Participants8 | EvidenceC | Evaluated |
| Rank04 | ModelMA | Score55.90% | Percentile57.14% | Participants8 | EvidenceC | Evaluated |
| Rank05 | ModelAC | Score48.20% | Percentile42.86% | Participants8 | EvidenceC | Evaluated |
| Rank06 | ModelAC | Score46.10% | Percentile28.57% | Participants8 | EvidenceC | Evaluated |
| Rank07 | ModelDE | Score38.00% | Percentile14.29% | Participants8 | EvidenceC | Evaluated |
| Rank08 | ModelAC | Score37.00% | Percentile0.00% | Participants8 | EvidenceC | Evaluated |
MCP-Mark Highlights
The leading models and scores on this benchmark.
MCP-Mark Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for MCP-Mark
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is MCP-Mark?
What MCP-Mark measures and how its scores work.
MCP-Mark is a benchmark for evaluating LLMs on Model Context Protocol (MCP) tool use.
It measures tool discovery, selection, invocation, and result interpretation, reported as a Score ratio.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- MCP-Mark
- Modality
- text
- Primary category
- agents
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- overall
LLM Stats. Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about MCP-Mark.
Which model scores highest on MCP-Mark?
Kimi K2.7 Code is currently ranked first with 81.10%.
What are the top three models on MCP-Mark?
The current leaders are Kimi K2.7 Code (81.10%), Qwen3.7 Max (60.80%), and Qwen3.7-Plus (58.70%).
Which MCP-Mark model has the lowest official input price?
Qwen3.6-35B-A3B has the lowest matched official input price at $0.25 input / $1.5 output per 1M tokens.
Which models are fastest among MCP-Mark results?
The fastest matched records are DeepSeek-V3.2 (97.00 tok/s via DeepInfra) and Kimi K2.6 (1.24 tok/s via Moonshot AI).
Does the highest score result prove overall model quality?
No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.