Benchmark shows V4 trailing Kimi and US frontier models, though gains raise hopes of another R1-style breakthrough
3-MIN READ3-MIN ListenXinmei ShenPublished: 4:51pm, 25 Apr 2026DeepSeek’s long-awaited flagship V4 model has fallen short of its domestic and US rivals, according to a new analysis, as the Chinese artificial intelligence firm struggles to replicate the market-shaking success of its earlier R1 release.
The company’s most advanced system, V4 Pro, ranked second among the world’s leading open-source models, behind Beijing-based Moonshot AI’s Kimi K2.6, benchmark firm Artificial Analysis said in a report on Friday.
While V4 Pro marked a clear improvement on its predecessor, V3.2, it still lagged top competitors.
The results highlight the challenges facing DeepSeek as China races to narrow the AI gap with the US, amid intensifying competition at home and abroad as well as ongoing constraints on computing power.
Still, analysts noted that V4 delivered meaningful technical progress.
