← Back to Research

Cross-Language Agent Framework Performance Benchmarking: Google ADK Go vs Python Agent Ecosystems

Faculty-advised research — team forming

→ Core contributors included as co-authors

Agent Framework Research

First systematic performance benchmark across multi-language agent development frameworks, covering latency, throughput, memory, and framework overhead under concurrent and sustained load conditions across Go and Python ecosystems.

Agent FrameworksBenchmarkingGoPythonLangGraphPerformanceMulti-Agent

The crisis

  • Agent framework performance directly determines deployment cost and latency for AI systems in resource-constrained environments
  • No standardized benchmark exists for comparing Go vs Python agent frameworks — practitioners make framework choices without empirical data
  • Framework overhead is invisible in current tooling — organizations optimize LLM calls while ignoring the framework layer beneath them

About this research

As agent development frameworks proliferate across languages and providers, practitioners lack empirical data for comparing them under realistic conditions. This work establishes the first cross-language benchmark covering Google ADK Go, ADK Python, LangGraph, and OpenAI Agents SDK across five scenarios — from single-agent baseline to high-concurrency sustained load. The benchmark isolates framework overhead from LLM API time, enabling fair comparison of the coordination layer itself.

Key findings

  • (In Progress)

Apply

Coordinate in Slack.

Apply Now →

Roles & contributors

Team

Lead — Go Implementation

Filled

Dishant Pandauria

Implements ADK Go benchmark, Go concurrency patterns, AWS execution.

Skills: Go, Google ADK, Benchmarking, AWS

Lead — Python + Analysis

Filled

Rashanjot Kaur

Implements LangGraph and OpenAI SDK benchmarks, paper writing coordination.

Skills: Python, LangGraph, OpenAI SDK, Research Design

Statistical Analysis

Filled

Siddhant Shah

Statistical analysis, chart generation, results validation.

Skills: Statistics, Python, Data Analysis

Faculty Advisor

Filled

Prof. Eugene Pinsky

Methodology review and publication venue selection.

Skills: Computer Science, Research Methodology