# MCP Atlas > A benchmark for tool-calling over Model Context Protocol integrations and external tools. Canonical page: https://benchlm.ai/benchmarks/mcpatlas - Category: [Agentic](/agentic) - Last updated: September 10, 2026 ## About MCP Atlas - Year: 2026 - Tasks: Tool-integrated agent tasks - Format: Interactive tool-calling evaluation - Difficulty: Advanced tool use - Paper: [Introducing GPT-5.4 mini and nano](https://openai.com/index/introducing-gpt-5-4-mini-and-nano/) OpenAI reports MCP Atlas as a tool-use benchmark that measures how well models work with MCP-backed systems and external tools. MCP Atlas is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula. ## Leaderboard (37 models) | Rank | Model | Creator | Score | |------|-------|---------|-------| | 1 | [Muse Spark 1.1](/models/muse-spark-1-1) | Meta | 88.1% | | 2 | [Claude Opus 5](/models/claude-opus-5) | Anthropic | 85.8% | | 3 | [Kimi K3](/models/kimi-k3) | Moonshot AI | 84.2% | | 4 | [Hy4 preview](/models/hy4-preview) | Tencent | 83.7% | | 5 | [Gemini 3.5 Flash](/models/gemini-3-5-flash) | Google | 83.6% | | 6 | [Claude Opus 4.8](/models/claude-opus-4-8) | Anthropic | 82.2% | | 7 | [Ornith-1.5-397B](/models/ornith-1-5-397b) | Ornith AI | 80% | | 8 | [Inkling-Small](/models/inkling-small) | Thinking Machines Lab | 79.6% | | 9 | [Claude Opus 4.7 (Adaptive)](/models/claude-opus-4-7-adaptive) | Anthropic | 77.3% | | 10 | [GLM-5.2](/models/glm-5-2) | Z.AI | 76.8% | | 11 | [Qwen3.7 Max](/models/qwen3-7-max) | Alibaba | 76.4% | | 12 | [Kimi K2.7 Code](/models/kimi-k2-7-code) | Moonshot AI | 76% | | 13 | [Muse Glimmer 30B](/models/muse-glimmer-30b) | Meta | 75.5% | | 14 | [GPT-5.5](/models/gpt-5-5) | OpenAI | 75.3% | | 15 | [MiniMax M3](/models/minimax-m3) | MiniMax | 74.2% | | 16 | [Inkling](/models/inkling) | Thinking Machines Lab | 74.1% | | 17 | [DeepSeek V4 Pro 0813](/models/deepseek-v4-pro-0813) | DeepSeek | 73.6% | | 18 | [Qwen3.7 Plus](/models/qwen3-7-plus) | Alibaba | 73.2% | | 19 | [GLM-5.1](/models/glm-5-1) | Z.AI | 71.8% | | 20 | [GPT-5.4](/models/gpt-5-4) | OpenAI | 70.6% | | 21 | [Ornith-1.5-35B-A3B](/models/ornith-1-5-35b-a3b) | Ornith AI | 70.2% | | 22 | [DeepSeek V4 Flash 0731](/models/deepseek-v4-flash-0731) | DeepSeek | 69% | | 23 | [Ling 3.0 Flash](/models/ling-3-0-flash) | InclusionAI | 65.5% | | 24 | [Qwen3.6-35B-A3B](/models/qwen3-6-35b-a3b) | Alibaba | 62.8% | | 25 | [Solar Pro 4](/models/solar-pro-4) | Upstage | 61.4% | | 26 | [Solar Open 2](/models/solar-open-2) | Upstage | 58.2% | | 27 | [GPT-5.4 mini](/models/gpt-5-4-mini) | OpenAI | 57.7% | | 28 | [GPT-5.4 nano](/models/gpt-5-4-nano) | OpenAI | 56.1% | | 29 | [Kimi K2.6](/models/kimi-2-6) | Moonshot AI | 55.9% | | 30 | [Ornith-1.5-9B](/models/ornith-1-5-9b) | Ornith AI | 54.2% | | 31 | [Qwen3.6 Plus](/models/qwen3-6-plus) | Alibaba | 48.2% | | 32 | [LLaDA2.2-flash](/models/llada2-2-flash) | InclusionAI | 46.2% | | 33 | [Qwen3.5 397B](/models/qwen3-5-397b) | Alibaba | 46.1% | | 34 | [LongCat-Flash-Lite-Sparse](/models/longcat-flash-lite-sparse) | Meituan | 45.6% | | 35 | [Claude Opus 4.5](/models/claude-opus-4-5) | Anthropic | 42.3% | | 36 | [GLM-5](/models/glm-5) | Z.AI | 31.1% | | 37 | [Kimi K2.5](/models/kimi-k2-5) | Moonshot AI | 29.5% | ## FAQ ### What does MCP Atlas measure? A benchmark for tool-calling over Model Context Protocol integrations and external tools. ### Which model scores highest on MCP Atlas? Muse Spark 1.1 by Meta currently leads with a score of 88.1% on MCP Atlas. ### How many models are evaluated on MCP Atlas? 37 AI models have been evaluated on MCP Atlas on BenchLM. ### Does MCP Atlas affect BenchLM's overall score? Not directly. MCP Atlas is still displayed on BenchLM for reference, but it is excluded from the weighted scoring formula. ## Compare Top Models on MCP Atlas - [Muse Spark 1.1 vs Claude Opus 5](/compare/claude-opus-5-vs-muse-spark-1-1) - [Claude Opus 5 vs Kimi K3](/compare/claude-opus-5-vs-kimi-k3) - [Kimi K3 vs Hy4 preview](/compare/hy4-preview-vs-kimi-k3) - [Hy4 preview vs Gemini 3.5 Flash](/compare/gemini-3-5-flash-vs-hy4-preview)