{"data":{"slug":"nvidia-h800","etype":"chip","name":"NVIDIA H800","cat":"GPU","subtitle":"The export-compliant H100 for China — and the chip DeepSeek trained V3 on.","summary":"The H800 kept the H100's FP8 tensor throughput but cut NVLink bandwidth to about 400 GB/s and sharply reduced FP64 to satisfy the 2022 export rules. DeepSeek's V3 technical report describes training on a 2,048-GPU H800 cluster, engineering around the interconnect limits — the chip is now shorthand for how constrained hardware shaped Chinese frontier labs. Banned by the October 2023 rules.","attrs":[{"g":"Overview","l":"Designer","v":"NVIDIA"},{"g":"Overview","l":"Architecture","v":"Hopper"},{"g":"Overview","l":"Introduced","v":"March 2023, China market"},{"g":"Overview","l":"Status","v":"Historical — barred by the October 2023 US export rules"},{"b":"Reported","g":"Physical","l":"Interconnect","v":"NVLink reduced to ~400 GB/s"},{"b":"Reported","g":"Compute","l":"FP64","v":"~1 TFLOPS (sharply reduced vs H100)"},{"b":"Reported","g":"Compute","l":"FP8/BF16 tensor","v":"Matches H100 published figures"}],"kf":{"node":"TSMC 4N","year":2023,"class":"GPU","tdp_w":700,"mem_bw":3.35,"memory":"80 GB HBM3","designer":"NVIDIA"},"aliases":["H800"],"ticker":null,"exchange":null,"hq":null,"founded":null,"website":null,"status":"Historical","region":null,"asof":"2026-01-15","sources":[{"url":"https://arxiv.org/abs/2412.19437","label":"DeepSeek-V3 technical report (H800 cluster)"}]},"meta":{"found":true,"count":1,"served_from":"site cache (no live upstream calls)","as_of":"2026-10-14T06:51:03+00:00","disclaimer":"Research and information only — not investment advice. Structured records carry basis labels (vendor spec / reported / estimate); market quotes are delayed/indicative.","attribution":"Records compiled from vendor documentation, official filings and reputable public reporting; market quotes via financial data providers. See /data-sources."}}