Arm has announced Arm Performix, a free performance analysis toolkit for modern agentic development workflows. It is designed to provide scalable performance for AI agents by unifying performance insights and optimization across the Arm compute platform. Performix enables developers and AI agents to understand, analyze, and optimize applications running on Arm-based cloud platforms through clear, deep performance insights. Microsoft, MongoDB, Redis, and SAP have shared support for the toolkit.
The launch follows Arm’s development of compute for AI workloads, including the Arm AGI CPU, with more than 1.25 billion Arm Neoverse cores shipped. Performix expands performance capabilities from silicon into the full software stack and integrates into automated, agent-driven workflows.
The shift to agentic AI has introduced complex workloads spanning multiple components. Performix addresses this by delivering system-wide visibility into inefficiencies on Arm-based infrastructure. It collects performance data directly from Arm-based silicon at runtime, covering metrics such as memory bandwidth, latency, cache efficiency, and CPU utilization, and converts them into guided, actionable insights.
The toolkit includes the Arm MCP Server, which allows developers and AI assistants to run Performix from tools such as GitHub Copilot, Kiro, Gemini, and Codex. Analysis can be triggered within the development environment, with results provided alongside code. This supports continuous, automated performance evaluation for tasks including migration and optimization.
In 2025, 50% of the CPU compute shipped to top hyperscalers was Arm-based, according to internal Arm estimates. Performix was developed with feedback from partners including Microsoft, MongoDB, Redis, and SAP.
“As developers and cloud architects increasingly seek to move existing x86 cloud workloads onto Arm, they need clear metrics and insights to help them identify performance bottlenecks efficiently. Arm Performix was developed with extensive feedback from our team to help simplify this migration,” said Pat Stemen, Vice President, Azure Hardware Systems and Infrastructure, Microsoft. “We look forward to seeing customers utilize Arm Performix to optimize their applications for Cobalt 200 virtual machines.”
“As a senior member of the performance engineering team at MongoDB, I was extremely impressed with the depth of analysis Arm Performix provides. The tool is approachable, and quick to set up and run,” said Jawwad Asghar, Senior Software Engineer at MongoDB. “I was able to generate a wealth of data about where to target optimization efforts across our code base. The CPU Microarchitecture analysis showed a clear front-end bottleneck and which functions were the most inefficient at different cache levels. Using the actionable insights Performix generated, we found a clear path to achieving a 10% performance boost on one of our key workloads in just a few days. The tool has been a big win for developer velocity in our performance tuning process.”
“Agentic AI workloads are driving unprecedented scale and complexity, making it critical to understand performance across the full stack,” said Filipe Oliveira, Performance Analysis and Optimization Lead, Redis. “We’re working with Arm to optimize Redis on Arm-based cloud infrastructure. Arm Performix brings system-wide visibility and actionable insights we need to identify inefficiencies quickly. Performix is an enabler to streamline and continuously optimize performance as we scale AI and data-intensive workloads.”
“Arm Performix streamlines performance analysis using intuitive visualizations and pre-configured recipes,” said Lars Hoemke, Head of HANA Core Performance, SAP. “With SAP HANA, Arm Performix allows us to identify code hotspots, drill down into issues, compare systems quickly and gain clear performance insights that save time.”
News Source : Arm
