MCPMark
MCPMark - GitHub MCP v0.30.3; Playwright responses truncated at 32K tokens.
Category
General Agents
Score type
percent
Scale
100%
higher is better
Models tested
11
What is MCPMark?
MCPMark is a benchmark for how well a model acts as a general-purpose agent. The official description says: "MCPMark - GitHub MCP v0.30.3; Playwright responses truncated at 32K tokens". Below we describe in simple words what benchmarks of this type generally examine.
What does it test?
MCPMark measures how well a model acts as a general-purpose agent. In detail, it checks:
- solving long, multi-step tasks with many intermediate decisions
- using tools correctly (browsers, terminals, APIs, calendars, other apps)
- adapting when something unexpected happens mid-task
- staying on goal over long task horizons instead of wandering off
How to read the score
The score is accuracy: the percentage of tasks answered or solved correctly. 100% means every task was solved perfectly; higher is better.
Official methodology note: MCPMark - GitHub MCP v0.30.3; Playwright responses truncated at 32K tokens.
Frequently asked questions
What is the MCPMark benchmark?
MCPMark is a benchmark for how well a model acts as a general-purpose agent. The official description says: 'MCPMark - GitHub MCP v0.30.3; Playwright responses truncated at 32K tokens'. Below we describe in simple words what benchmarks of this type generally examine.
What does MCPMark test?
MCPMark mainly measures how well a model acts as a general-purpose agent. Concretely, it checks: solving long, multi-step tasks with many intermediate decisions; using tools correctly (browsers, terminals, APIs, calendars, other apps); adapting when something unexpected happens mid-task.
How is the MCPMark score calculated?
The score is accuracy: the percentage of tasks answered or solved correctly. 100% means every task was solved perfectly; higher is better. The declared maximum score of this benchmark is 100.0.
Where do I get MCPMark data and scores?
The official benchmark page/reference is linked on this page. Every score on Open LLM Radar is also available as open data: download the scores JSON (http://open-llm-radar-api:8000/api/v1/benchmarks/mcpmark/scores) or pull it directly from the Open LLM Radar API with any HTTP client.
Benchmark data & downloads
Where to get, inspect and reuse the data behind this benchmark.
Benchmark dataset
The question/task data itself — download and evaluate locally
Benchmark metadata (JSON)
This page's benchmark record from the Open LLM Radar API
All model scores (JSON download)
Full score table as JSON; paginate with limit/offset params
Open LLM Radar API docs
Swagger UI to explore and reuse all radar API endpoints
Model scores 11 models tested
| # | Model | Score | Date | Source | Verified |
|---|---|---|---|---|---|
| 1 |
Kimi K3
Moonshot AI
|
94.5%
|
18.08.2026 | Report | Verified |
| 2 |
Kimi K2.6
Moonshot AI
|
55.9%
|
23.08.2026 | Report | Verified |
| 3 |
Qwen3.6-35B-A3B
Qwen Team (Alibaba Cloud)
|
37%
|
16.04.2026 | Qwen Team (official model card) | Verified |
| 4 |
Qwen3.5-27B
Qwen Team (Alibaba Cloud)
|
36.3%
|
16.04.2026 | Qwen Team (official model card - comparison) | Unverified |
| 5 |
Kimi K2.5
Moonshot AI
|
29.5%
|
23.08.2026 | Report | Verified |
| 6 |
Qwen3.5-35B-A3B
Qwen Team (Alibaba Cloud)
|
27%
|
16.04.2026 | Qwen Team (official model card - comparison) | Unverified |
| 7 |
Gemma4-31B-it
Google
|
18.1%
|
16.04.2026 | Qwen Team (official model card - comparison) | Unverified |
| 8 |
Gemma4-26B-A4B
Google
|
14.2%
|
16.04.2026 | Qwen Team (official model card - comparison) | Unverified |
| 9 |
Spark-X2.5-4B
SparkLLM (XHToken)
|
14.2%
|
— | Report | Verified |
| 10 |
Qwen3.5-4B
Qwen
|
12.5%
|
— | Report | Verified |
| 11 |
Spark-X2.5-1.7B
SparkLLM (XHToken)
|
2.3%
|
— | Report | Verified |