Artificial Analysis says China’s top model is still 3–9 months behind the US lead
ArtificialAnlys · x · 2026-08-04
- Artificial Analysis chief Micah Hill-Smith said on Bloomberg TV that over nearly three years of its Intelligence Index, the leading model from a China-based lab has generally trailed the leading US-lab model by 3–9 months.
- He said Moonshot’s Kimi K3 is currently the top Chinese model and ranks 4th overall on the index.
- He also highlighted DeepSeek’s updated V4 Flash 0731, which launched last week and stands out on the Intelligence vs. Cost per Task chart.
- Another takeaway: the open-weights frontier has been dominated by China-based labs for the past two years.
More from Models
- OpenAI hires the creator of WebRTC as its GPT-Live voice system gets a deep dive — bookwormengr · 2026-08-04
- Code Arena WebDev puts four open-weight Chinese models near the frontier — floriandotorg · 2026-08-04
- GPT-5.6 Misspells Email Address, Then Hallucinates a Post-Hoc Excuse — WolframRvnwlf · 2026-08-04
- Palantir says Nvidia’s Nemotron Ultra beat frontier models on real customer tasks — eliano · 2026-08-04
- Gemma 4 26B runs 24 concurrent users on a single RTX 4090 with llama.cpp — DynamicWebPaige · 2026-08-04
- Users say frontier models are getting harder to communicate with as jargon and verbosity rise — StewartalsopIII · 2026-08-04