Huawei's HiZQ is real HBM; DeepSeek's designs all target bandwidth, not storage
zephyr_z9 · x · 2026-09-22
A technical thread notes that Huawei's HiZQ is genuine HBM (their cHBM variant), not a substitute. The discussion highlights that DeepSeek's architectural choices (MoE, sparse attention) primarily cut bandwidth rather than storage, since bandwidth is the bigger and more expensive constraint — with Engram representing the pinnacle of this bandwidth-first innovation path.
Related event: DeepSeek Engram Is Semantic Retrieval, Can Cut HBM Needs 40-50%(4 posts)→
More from Infra
- Photoroom reclaims ~1,000 idle GPU hours a week with SkyPilot across Slurm and Kubernetes — skypilot_org · 2026-11-12
- Cloudflare Talk: Swapping Some Agent LLM Calls for Fast Models Like Jev Gives Big Speedups — michellechen · 2026-11-12
- Cloudflare Trains Its Own Clef and Clairic Models to Complete Its Agent Stack — michellechen · 2026-11-12
- AWS open-sources 23 production-ready GenAI assets, cutting build time from months to weeks — AWS ML Blog · 2026-11-12
- Sixth Street CEO: ~$200B of Chips Sitting in Warehouses as Data Centers Lag Behind — SumitGup · 2026-11-12
- AWS CEO Matt Garman: Space Data Centers Are 10-20 Years Away, Likely Not Economically Feasible — rohanpaul_ai · 2026-11-12