InferenceX is SemiAnalysis's open, vendor-neutral inference benchmark that runs nightly on NVIDIA and AMD accelerators. It reports token throughput, latency, and cost per token for frontier models on real hardware.
Results are produced by public GitHub Actions workflows and published on a dashboard so operators can compare chips, frameworks, and configurations without relying on vendor claims.

InferenceX continuously benchmarks LLM inference throughput, latency, and cost across GPU vendors.


SiliconMark independently benchmarks how physical GPU clusters perform in real workloads.
SiliconMark independently benchmarks how physical GPU clusters perform in real workloads.




Unified production inference platform with gateway, router, cache, observability, evaluation, fine-tuning, and custom GPU hosting.


AI inference platform delivering serverless and dedicated GPU inference with top-ranked speed and price-performance.


Inference Neocloud delivers dedicated ASIC racks for high-speed AI inference workloads.


Baseten Dedicated Inference is a high-performance inference platform for deploying, serving, and scaling custom and open-source AI models in production.


Baseten Inference Stack for high performance with custom kernels, decoding, caching on optimized infra.


Crusoe Managed Inference deploys and scales production LLM workloads with breakthrough speed and throughput.


IonRouter is a high-throughput, low-cost inference API and drop-in OpenRouter alternative powered by Cumulus Labs' IonAttention engine.



Open-source USD price per GPU-hour across H100, H200, B200, and B300, refreshed every 15 minutes.



Cloudflare AI: serverless inference, agents, RAG, and secure runtime primitives on the global edge.


Open LLM Leaderboard ranks open language models on reproducible public evaluation tasks.


SiliconPriceIQ forecasts and normalizes GPU rental and resale prices across vendors and regions.


Weave Router is a drop-in proxy that routes each inference request to the lowest-cost model that answers at frontier parity.


An integrated compute, training, inference, and sandbox stack for enterprises to train and deploy their own AI models.


Cloud GPU Rental Price Index compiles monthly cloud GPU rental prices across providers and models.


Open-source inference serving platform for deploying AI models from any framework on CPUs and GPUs.


Pipeshift's platform provides infrastructure for fine-tuning, inference, and orchestration of open-source AI models.
MLPerf is an industry-standard suite that measures AI training and inference system performance.

Unified production inference platform with gateway, router, cache, observability, evaluation, fine-tuning, and custom GPU hosting.

AI inference platform delivering serverless and dedicated GPU inference with top-ranked speed and price-performance.

Inference Neocloud delivers dedicated ASIC racks for high-speed AI inference workloads.

Baseten Dedicated Inference is a high-performance inference platform for deploying, serving, and scaling custom and open-source AI models in production.

Baseten Inference Stack for high performance with custom kernels, decoding, caching on optimized infra.

Crusoe Managed Inference deploys and scales production LLM workloads with breakthrough speed and throughput.

IonRouter is a high-throughput, low-cost inference API and drop-in OpenRouter alternative powered by Cumulus Labs' IonAttention engine.

High-performance deep learning inference SDK for optimizing trained models on NVIDIA GPUs.

Open-source USD price per GPU-hour across H100, H200, B200, and B300, refreshed every 15 minutes.

KV-cache transfer engine that prefills small models and decodes large ones.

Cloudflare AI: serverless inference, agents, RAG, and secure runtime primitives on the global edge.

Open LLM Leaderboard ranks open language models on reproducible public evaluation tasks.

SiliconPriceIQ forecasts and normalizes GPU rental and resale prices across vendors and regions.

Weave Router is a drop-in proxy that routes each inference request to the lowest-cost model that answers at frontier parity.

An integrated compute, training, inference, and sandbox stack for enterprises to train and deploy their own AI models.

Cloud GPU Rental Price Index compiles monthly cloud GPU rental prices across providers and models.

Open-source inference serving platform for deploying AI models from any framework on CPUs and GPUs.

Pipeshift's platform provides infrastructure for fine-tuning, inference, and orchestration of open-source AI models.
