From 480cfc3c980b50e3072fd20c71b097db897a0a69 Mon Sep 17 00:00:00 2001 From: functionstackx <47992694+functionstackx@users.noreply.github.com> Date: Fri, 20 Mar 2026 19:46:24 -0400 Subject: [PATCH 1/3] Update page-content.tsx --- packages/app/src/components/page-content.tsx | 30 +++++++------------- 1 file changed, 11 insertions(+), 19 deletions(-) diff --git a/packages/app/src/components/page-content.tsx b/packages/app/src/components/page-content.tsx index 0f6fc4aa..5b80be72 100644 --- a/packages/app/src/components/page-content.tsx +++ b/packages/app/src/components/page-content.tsx @@ -256,31 +256,23 @@ export function PageContent({ initialTab = 'inference' }: { initialTab?: string

- LLM inference performance is a major concern in providing AI services, but - accurate performance analysis remains elusive. + Open Source Continuous Inference Standard and Research Platform trusted by Operators of Trillion Dollar Token Factories such as OpenAI, Oracle, Microsoft, etc.

- The fast cadence of software development and model releases makes comparing - performance between setups difficult. Existing performance benchmarks quickly - become obsolete because they are static, and participants often game the - benchmarks with unrealistic, highly specific configurations. + In the AI world, software development and model releases move at the speed of light. + Existing performance benchmarks become obsolete because they are static, and participants + often submit software images purpose built for the benchmark itself.

- InferenceX (formerly InferenceMAX) addresses these issues by - benchmarking popular models on major hardware platforms, updating benchmarks as - new software versions are released. -

-

- For each model and hardware combination, InferenceX sweeps - through different tensor parallel sizes and maximum concurrent requests, - presenting a throughput vs. latency graph for a complete picture. In terms of - software configurations, we ensure they are broadly applicable across different - serving scenarios, and we open-source the repo to encourage community - contributions. + InferenceX (formerly InferenceMAX) is our independent, reproducible + benchmark which addresses these issues by continously benchmarking inference software + across an wide range of AI accelerators that is acutally available to the the ML community. + We continously update the benchmarks to capture the speed of progress.

- We hope InferenceX provides the community with up-to-date and - realistic LLM inference performance insights. Learn more in our articles:{' '} + Our open data & insights is widely adopted by the ML community, capacity planning strategy teams at trillion dollar token factories & AI Labs + & at multiple billion dollar NeoClouds. + Learn more in our articles:{' '} Date: Fri, 20 Mar 2026 19:50:41 -0400 Subject: [PATCH 2/3] Update page-content.tsx --- packages/app/src/components/page-content.tsx | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/packages/app/src/components/page-content.tsx b/packages/app/src/components/page-content.tsx index 5b80be72..1593e771 100644 --- a/packages/app/src/components/page-content.tsx +++ b/packages/app/src/components/page-content.tsx @@ -256,12 +256,13 @@ export function PageContent({ initialTab = 'inference' }: { initialTab?: string

- Open Source Continuous Inference Standard and Research Platform trusted by Operators of Trillion Dollar Token Factories such as OpenAI, Oracle, Microsoft, etc. + Open Source Continuous Inference Standard and Research Platform trusted by Operators of Trillion Dollar Token Factories

- In the AI world, software development and model releases move at the speed of light. - Existing performance benchmarks become obsolete because they are static, and participants - often submit software images purpose built for the benchmark itself. + As the world progresses exponentially towards AGI, software development and model releases move at the speed of light. + Existing benchmarks rapidly become obsolete due to their static nature, and participants + often submit software images purpose built for the benchmark itself which do not reflect + real world performance

InferenceX (formerly InferenceMAX) is our independent, reproducible From c0f42d6f2990c6d98242618bd18b98ce616a9217 Mon Sep 17 00:00:00 2001 From: functionstackx <47992694+functionstackx@users.noreply.github.com> Date: Fri, 20 Mar 2026 19:53:58 -0400 Subject: [PATCH 3/3] fix: apply formatting fixes to page-content.tsx Co-Authored-By: Claude Opus 4.6 (1M context) --- packages/app/src/components/page-content.tsx | 26 +++++++++++--------- 1 file changed, 14 insertions(+), 12 deletions(-) diff --git a/packages/app/src/components/page-content.tsx b/packages/app/src/components/page-content.tsx index 1593e771..d72d2489 100644 --- a/packages/app/src/components/page-content.tsx +++ b/packages/app/src/components/page-content.tsx @@ -256,24 +256,26 @@ export function PageContent({ initialTab = 'inference' }: { initialTab?: string

- Open Source Continuous Inference Standard and Research Platform trusted by Operators of Trillion Dollar Token Factories + Open Source Continuous Inference Standard and Research Platform trusted by + Operators of Trillion Dollar Token Factories

- As the world progresses exponentially towards AGI, software development and model releases move at the speed of light. - Existing benchmarks rapidly become obsolete due to their static nature, and participants - often submit software images purpose built for the benchmark itself which do not reflect - real world performance + As the world progresses exponentially towards AGI, software development and model + releases move at the speed of light. Existing benchmarks rapidly become obsolete + due to their static nature, and participants often submit software images purpose + built for the benchmark itself which do not reflect real world performance

- InferenceX (formerly InferenceMAX) is our independent, reproducible - benchmark which addresses these issues by continously benchmarking inference software - across an wide range of AI accelerators that is acutally available to the the ML community. - We continously update the benchmarks to capture the speed of progress. + InferenceX (formerly InferenceMAX) is our independent, + reproducible benchmark which addresses these issues by continously benchmarking + inference software across an wide range of AI accelerators that is acutally + available to the the ML community. We continously update the benchmarks to capture + the speed of progress.

- Our open data & insights is widely adopted by the ML community, capacity planning strategy teams at trillion dollar token factories & AI Labs - & at multiple billion dollar NeoClouds. - Learn more in our articles:{' '} + Our open data & insights is widely adopted by the ML community, capacity planning + strategy teams at trillion dollar token factories & AI Labs & at multiple billion + dollar NeoClouds. Learn more in our articles:{' '}