Quick facts
- Best for
- InferenceX - Open Continuous LLM Inference Benchmarking
- Pricing
- Free
- Editor rating
- 4.5 / 5
- Community saves
- 0
About InferenceX
InferenceX is an open-source continuous inference benchmark research platform from SemiAnalysisAI. The repository metadata describes benchmark work for LLM inference across model workloads and accelerator stacks, with topics including AI, AMD, NVIDIA, CUDA, ROCm, GB200, PyTorch, vLLM, SGLang, and benchmarking.
Pros
- Track continuous LLM inference benchmark research in a public repository
- Study benchmark context across NVIDIA, AMD, CUDA, ROCm, vLLM, and SGLang topics
- Review scripts and project metadata for model-serving performance experiments
- Compare inference claims against source-visible benchmark work
- Use Apache-2.0 licensed materials as a starting point for internal research
Cons
Pricing
Open source
$0
- • Public source code is available
- • No required license fee is listed in the GitHub repository metadata
- • Hosting, storage, model APIs, cloud services, or commercial support can add costs