Skip to content

Benchmark

MCP Atlas

Agentic benchmarks measure how well models carry out tasks that take several steps. Scores are percentages. Higher is better.

Models and systems measured
46
Scores recorded
95
Sources
2
IQ weight
1.56

Score distribution

71.6%
Mean
75.3%
Median
11.6%
Std dev
57.7%
10th percentile
83.7%
90th percentile
24.6%
Min
24.6%89.2%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.56 for chat models, measured over 45 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#Model or systemScore
51Qwen3.6 Plus· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim74.1%
52Qwen3.6 PlusQwen official release table · thinking mode unless benchmark requires otherwise · public set · Gemini 2.5 Pro judge· Reported by Alibaba Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.74.1%
53Qwen3.7 Plus· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim73.2%
54Qwen3.7 PlusQwen3.7-Plus release · public set · Gemini 2.5 Pro judge· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.73.2%
55GLM 5.1· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim71.8%
56GLM 5.1GLM-5.1 official release table · public set· Reported by Z.ai· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.71.8%
57Claude Sonnet 4.6Gemini 3.5 Flash Model Card · May 2026· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.69.5%
58Kimi K2.6Kimi K2.7 Code official comparison table · Kimi K2.6 · Kimi Code CLI · thinking · temp 1.0 · top_p .95 · 262,144 context · 100 tool-call budget · 32K max tokens/step · avg@3· Reported by Moonshot AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.69.4%
59Gemini 3.1 Pro Preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim69.2%
60Gemini 3.1 Pro Previewpublic· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.69.2%
61DeepSeek V4 FlashDeepSeek V4 official model card · Think Max· Reported by DeepSeek· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.69%
62GLM 5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim67.8%
63GLM 5GLM-5 official release table · public set· Reported by Z.ai· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.67.8%
64DeepSeek V4 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim67.4%
65DeepSeek V4 FlashDeepSeek V4 official model card · Think High· Reported by DeepSeek· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.67.4%
66GPT 5.4· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim67.2%
67GPT 5.4OpenAI GPT-5.4 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.67.2%
68Ling 3.0 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim65.5%
69DeepSeek V4 FlashDeepSeek V4 official model card · Non-Think· Reported by DeepSeek· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.64%
70Qwen3.6 35B A3B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim62.8%
71Qwen3.6 FlashQwen3.6-35B-A3B official model card · hosted API name Qwen3.6-Flash· Reported by Alibaba Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.62.8%
72Claude Opus 4.6· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim62.7%
73Claude Opus 4.6Anthropic Opus 4.6 launch page · adaptive thinking · max effort unless noted · high effort· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.62.7%
74Claude Opus 4.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim62.3%
75Claude Opus 4.5no extended thinking · 200K context · default sampling· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.62.3%
76Claude Opus 4.5Opus 4.6 System Card capability summary · adaptive thinking · max effort · default sampling · 5 trials unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.62.3%
77Gemini 3 FlashGemini 3.5 Flash Model Card · May 2026· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.62%
78Claude Sonnet 4.6· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim61.3%
79Claude Sonnet 4.6max effort· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.61.3%
80GPT 5.2· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim60.6%
81GPT 5.2OpenAI GPT-5.4 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.60.6%
82GPT 5.2Opus 4.6 System Card capability summary · adaptive thinking · max effort · default sampling · 5 trials unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.60.6%
83GPT 5.2Scale MCP-Atlas · xhigh· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.60.6%
84Claude Opus 4.6Opus 4.6 System Card capability summary · adaptive thinking · max effort · default sampling · 5 trials unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.59.5%
85GPT-5.4 mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim57.7%
86GPT-5.4 minixhigh · published setting· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.57.7%
87Gemini 3 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim57.4%
88Gemini 3 FlashScale AI MCP Atlas· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.57.4%
89GPT 5.4 Nano· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim56.1%
90GPT 5.4 Nanoxhigh · published setting· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.56.1%
91Gemini 3 Pro PreviewOpus 4.6 System Card capability summary · adaptive thinking · max effort · default sampling · 5 trials unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.54.1%
92Claude Sonnet 4.5Opus 4.6 System Card capability summary · adaptive thinking · max effort · default sampling · 5 trials unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.43.8%
93Nova 2 Lite· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim24.6%
94Nova 2 LiteAmazon Nova 2 Lite official launch chart / technical report · blind evaluation · pass rate, v1· Reported by Amazon Web Services, Amazon AGI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.24.6%

Showing 51–94 of 94 results.

Not ranked

The source marks these results as descriptive, not capability scores, so we list them by name instead of ranking them.

Model or systemScore
Muse Spark (version unspecified)Meta Model API · highest effort · version unspecified · Scale AI MCP Atlas· Reported by Meta, Meta Superintelligence Labs· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.82.2%

Data credits: Data by LLM Stats · BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links.