InferenceXbySemiAnalysis logo
HomeAgentXNEWOverviewDashboardComparisonsArticlesAbout
Star1,583中文

Articles

Articles on agentic inference, AgentX results, chip performance, and ML infrastructure.

New to the terminology? Browse the AI inference glossary.

Allagenticagentsagentxamdannouncementasicb200b300benchmarkcanndeepseekdisaggdynamofp4fp8gb200gb300glm5gpuh100h200huaweiinferencekimilatencymi355xminimaxnvfp4nvidianvl72openaiqwenrocmrubinsglangthroughputtilerttrtllmvllmwide-ep
May 26, 2026·11 min read

B200 NVFP4 vs H200 INT4 on Kimi K2.5/K2.6: Up to 2.95x Better Performance per Dollar

On vLLM 8K/1K the NVFP4 path on B200 is 2.71x–2.95x cheaper per million tokens than H200 INT4 across the entire 30–90 tok/s/user serving band, and 2.45x–2.74x cheaper than B200 INT4 on the same silicon. Both factors decompose cleanly into B200's HBM bandwidth, HBM capacity, and NVFP4 tensor cores

benchmarkgpuinferencekiminvidiab200h200vllmnvfp4
SemiAnalysis logo

Continuous open-source agentic inference benchmarking. Real-world, reproducible, auditable performance data trusted by trillion dollar AI infrastructure operators like OpenAI, Meta, Oracle, Microsoft, etc.

SemiAnalysisMain SiteNewsletterAbout
LegalLand AcknowledgementPrivacy PolicyCookie Policy
ContributeBenchmarksAgentX HarnessVisualization
More
SupportersAgentXTelemetryArticlesAPI ReferenceChip ReliabilityPerformance per DollarModel ArchitecturesAI Inference GlossaryChip Specs & PricingGPU RankingsModel on GPU Results

If this data helps your work, consider starring us on GitHub or sharing with your network.

© 2026 semianalysis.com. All rights reserved.