SemiAnalysis launched InferenceMAX, an open-source nightly benchmark comparing GPU inference performance across NVIDIA (H100, H200, B200, GB200 NVL72) and AMD (MI300X, MI325X, MI355X). Endorsed by Jensen Huang, Lisa Su, OpenAI, and Microsoft. First multi-vendor benchmark with TCO and power efficiency metrics.
91%
Overall ConfidenceBased on 5 weighted dimensions
Confidence Breakdown
Authority
30%95Tier 1
Data Quality
25%100Level 5
Recency
20%4091-180 days
Corroboration
15%1003 sources
Specificity
10%100Level 4
Data Points(3)
InferenceMAX runs nightly benchmarks across 7 GPU types using vLLM, SGLang, TRT-LLM
Oct 9, 2025Expanding to Google TPU and AWS Trainium within 2 months for true multi-vendor comparison
Oct 9, 2025Benchmark includes TCO per million tokens and tokens per provisioned MW metrics
Oct 9, 2025