PulseAugur
EN
LIVE 08:30:54

DeepSeek V4 and Huawei Ascend 950DT cut AI inference costs by 75%

DeepSeek V4, an AI model, has been co-designed with Huawei's Ascend 950DT AI accelerator. This collaboration has reportedly led to a significant 75% reduction in AI inference costs. The analysis was conducted by the research firm SemiAnalysis. AI

IMPACT This co-design highlights potential for significant cost savings in AI inference, which could accelerate adoption of advanced models.

RANK_REASON The cluster details a co-design effort between an AI model and hardware, with a specific cost reduction metric reported by an analysis firm. [lever_c_demoted from research: ic=1 ai=1.0]

Read on Pandaily →

AI-generated summary · Google Gemini · from 2 sources. How we write summaries →

DeepSeek V4 and Huawei Ascend 950DT cut AI inference costs by 75%

COVERAGE [2]

  1. Pandaily TIER_1 English(EN) · [email protected] (Pandaily) ·

    DeepSeek V4 and Huawei Ascend 950DT: The Co-Designed Chip That Cut AI Inference Costs by 75%

    A groundbreaking trace-level analysis by Wall Street research firm SemiAnalysis has revealed that DeepSeek V4 and Huawei Ascend 950DT AI accelerator were co-des...

  2. Mastodon — mastodon.social TIER_1 English(EN) · [email protected] ·

    DeepSeek V4 and Huawei Ascend 950DT AI accelerator were co-designed from scratch, cutting inference costs by 75% to just 0.20 yuan per million tokens. SemiAnaly

    DeepSeek V4 and Huawei Ascend 950DT AI accelerator were co-designed from scratch, cutting inference costs by 75% to just 0.20 yuan per million tokens. SemiAnalysis found Huawei's CANN 8.5 stack and the unique dual-die 950DT architecture were built specifically for DeepSeek infere…