PulseAugur
EN
LIVE 06:09:32
한국어(KO) Brian Armstrong (@brian_armstrong) 앞으로 지능 수요는 사실상 무한하지만, 12~18개월 내에 워크로드의 80%는 99% 더 저렴한 모델에서 돌고, 나머지 20%만 최신 세대 모델을 사용할 것이라는 전망입니다. 실무에서는 비용 최적화와 고성능 모델 사용의 분리

Coinbase CEO: 80% of AI workloads to use cheaper models in 18 months

By PulseAugur Editorial · [1 sources] ·

Brian Armstrong, CEO of Coinbase, predicts that within 12 to 18 months, 80% of AI workloads will run on significantly cheaper models, while only 20% will utilize the latest generation. This suggests a growing separation between cost optimization and the use of high-performance AI models in practical applications. AI

IMPACT Predicts a significant shift towards cost-effective AI models, potentially impacting development and deployment strategies for AI applications.

RANK_REASON The cluster contains an opinion/forecast from a prominent figure in the tech industry about the future of AI model usage and cost.

Read on Mastodon — fosstodon.org →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

COVERAGE [1]

  1. Mastodon — fosstodon.org TIER_1 한국어(KO) · [email protected] ·

    Brian Armstrong (@brian_armstrong) predicts that while demand for intelligence is virtually infinite, 80% of workloads will run on 99% cheaper models within 12-18 months, with only the remaining 20% using the latest generation models. In practice, this means separating cost optimization from the use of high-performance models.

    Brian Armstrong (@brian_armstrong) 앞으로 지능 수요는 사실상 무한하지만, 12~18개월 내에 워크로드의 80%는 99% 더 저렴한 모델에서 돌고, 나머지 20%만 최신 세대 모델을 사용할 것이라는 전망입니다. 실무에서는 비용 최적화와 고성능 모델 사용의 분리가 더 뚜렷해질 가능성을 시사합니다. https:// x.com/brian_armstrong/status/2 063782620815876515 # llm # modelcost # inference # ai # for…

RELATED ENTITIES

RELATED TOPICS