English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The carbon footprint associated with large language models (LLMs) is asignificant concern, encompassing emissions from their training, inference,experimentation, and storage processes, including operational and embodiedcarbon emissions. An essential aspect is accurately estimating the carbonimpact of emerging LLMs even before their training, which heavily relies on GPUusage. Existing studies have reported the carbon footprint of LLM training, butonly one tool, mlco2, can predict the carbon footprint of new neural networksprior to physical training. However, mlco2 has several serious limitations. Itcannot extend its estimation to dense or mixture-of-experts (MoE) LLMs,disregards critical architectural parameters, focuses solely on GPUs, andcannot model embodied carbon footprints. Addressing these gaps, we introduce\textit{\carb}, an end-to-end carbon footprint projection model designed forboth dense and MoE LLMs. Compared to mlco2, \carb~significantly enhances theaccuracy of carbon footprint estimations for various LLMs. The source code isreleased at \url{https://github.com/SotaroKaneda/MLCarbon}.

LLMCarbon: Modeling the end-to-end Carbon Footprint of Large Language Models