SemiAnalysisAI/InferenceX

Star SemiAnalysisAI / InferenceX Open Source Continuous Inference Benchmarking Qwen3.5, DeepSeek, GPTOSS - GB200 NVL72 vs MI355X vs B200 vs GB300 NVL72 vs H100 & soon™ TPUv6e/v7/Trainium2/3

Python1.7K stars292 forks+6 today

View on GitHubShare on X

Topics

#pytorch#sglang#vllm#amd#cuda#nvidia#rocm#deepseek#kimi#agentic#agentic-ai#benchmarking#benchmarks#gpu#inference#llm-inference#llm-serving#mlops#qwen#tensorrt-llm

What it does

InferenceX™ is an open-source benchmarking tool that continuously evaluates the performance of popular inference frameworks in real-time. It addresses the rapid advancements in AI software by providing up-to-date performance metrics, ensuring developers can rely on accurate data for optimization.

Star history

2026-08-23 → 2026-09-12+246 stars

Tracking

Last trending: 2026-02-21

Creator kit

Hook

Discover InferenceX™: the tool that keeps AI inference benchmarks up-to-date with real-time performance tracking!

Content angles

Create a tutorial on how to set up and use InferenceX™ for benchmarking AI models.
Write a comparison article on the performance of different inference frameworks using InferenceX™ results.
Develop a video series showcasing the evolution of AI inference performance over time with InferenceX™ data.

Who should care

AI developers, data scientists, and machine learning engineers interested in performance optimization.