technology
China · Best AI Model Capability (Epoch Capabilities Index)
Latest
157.42ECI score
Methodology
## What this measures
The Epoch Capabilities Index score of the most capable AI model released so far by a developer in China. ECI combines results on about 40 benchmarks into one scale per model. Each month's value is the highest ECI score among models released on or before the last day of that month (the current month reads to date). A model counts for a country when Epoch lists an organisation from that country among its developers. The series starts in January 2024: Epoch had scored one Chinese model by June 2023 and eleven by December 2023, so earlier months would show which models were scored rather than the frontier.
## Why it matters
It is the most direct public read of where China stands in the race to build the most capable AI models. Set beside the United States (ai.capability.eci_frontier.usa) it shows who leads and by how much.
## How to read it
Higher means a more capable best model. The line steps up when a new model beats the old best and stays flat in between; it never falls. Read the gap to the other country in ai.capability.eci_frontier_gap.usa_chn.
## What it does not say
It counts only models Epoch has scored, so a strong model it has not yet benchmarked is missing until it is. Benchmarks favour some skills over others, and a developer's country is where its organisation is based, not where its models are used. It says nothing about adoption, cost or compute.
## Source
Epoch AI, Epoch Capabilities Index (epoch.ai/eci). CC BY 4.0. Rewritten weekly, because Epoch re-estimates every score when it adds benchmarks.
Read this signal inside the lattice.
Palanor weighs every signal against the world your organization is watching. Methodology is public; integration is the platform.
See it in Palanor →