Vibeleaderboard
← All Intel
Intel / post

Korean labs lead non-US, non-China models on Artificial Analysis index

Source
x.com
Date
Artificial Analysis@ArtificialAnlys

South Korea has once again established a clear #3 position in the global AI race, trailing only the United States and China. The depth and vibrancy of Korea’s AI ecosystem is supported by strong domestic talent, public investment, and the incentives created by Korea’s Sovereign AI Foundation Model project A number of South Korean labs, including Motif Technologies, Upstage, SK Telecom, and LG AI Research, have now scored above 30 on the Artificial Analysis Intelligence Index. This concentration of competitive model developers highlights the growing density of Korea’s domestic AI ecosystem. Among them, Motif Technologies and Upstage have models scoring above 40, making them the highest-scoring models on the Intelligence Index developed outside the United States and China. Leading models from each lab include: ➤ Motif Technologies’ Motif 3 scored 47 on the Intelligence Index (314B total parameters, 13B active; open weights). ➤ Upstage’s Solar Pro 4 scored 42 on the Intelligence Index. It is a proprietary reasoning model. ➤ SK Telecom’s A.X K2 scored 35 on the Intelligence Index (692B total parameters, 33B active; open weights). ➤ LG AI Research’s K-EXAONE 2.0 scored 31 on the…

Read the full post on X

Context

Artificial Analysis's claim that South Korea holds third place in AI rests on its own : Korean models are the highest scorers on its Intelligence Index from anywhere outside the United States and China. In an August 25 post, the evaluator lists Motif Technologies' Motif 3 at 47 and Upstage's Solar Pro 4 at 42, followed by SK Telecom's A.X K2 at 35 and LG AI Research's K-EXAONE 2.0 at 31. Its point is depth as well as height: four separate Korean developers now have models above 30. Three of the four are open-weights models anyone can download and run, while Solar Pro 4 is a proprietary .

Third place is a long way from second. The accompanying chart puts the leading US model at 63 and the leading Chinese model at 60, against Korea's 47, while the next country shown, France, peaks at 30 with Mistral AI's Mistral Medium 3.5. So Korea's position is a clear lead over every other country on the chart outside the two leaders, not closeness to them. The index itself combines nine evaluations, among them Terminal-Bench, SciCode, Humanity's Last Exam and GPQA Diamond, so a ranking of countries here is a ranking of their best models on those tests.

The reasons Artificial Analysis offers, domestic talent, public investment and the incentives of Korea's Sovereign AI Foundation Model project, are its interpretation rather than something the scores show. The scores establish that several Korean labs can build models that score competitively on these tests; they do not measure the size of Korea's AI industry, its compute or its adoption.

Terms in this piece · Glossary
  • benchmark — A standard public test set for comparing AI models — the shared scoreboards behind every "model X beats model Y" claim.
  • reasoning model — A model trained to think — generating extended internal reasoning before answering — trading time and tokens for accuracy on hard problems.
  • open weights — A model whose trained parameters are published for anyone to download and run — unlike API-only models you can access but never possess.
More from Artificial Analysis
Recommended reads
Comments

Checking sign-in…

Loading comments…