DeepSeek narrows US-China AI gap to 3%, says Bloomberg Intelligence
Chinese AI models are now within 3% of their US rivals. Bloomberg Intelligence reported the narrowing gap on October 5.
DeepSeek’s V4.1 Flash drove the shift. The model ranked sixth on LiveBench with 81.1 points, close to Anthropic’s top score of 83.4.
The gap was 9% in May and 15% earlier this year. The pace of catch-up has surprised analysts.
Senior analyst Robert Lee said DeepSeek performs at levels comparable to leading US models. China has combined low costs with rising quality.
The finding raises questions about how long US dominance will last. The AI race is tightening fast.
DeepSeek is a Chinese AI lab based in Hangzhou. Its models are known for low cost and strong performance. The V4.1 Flash model is open for developers to use. US labs still lead on the most complex tasks, analysts said. But the cost gap favours Chinese models. The findings were based on benchmark scores and pricing data.
Related stories
View all →
Eloelo launches Dolphin AI video tool at FICCI FRAMES
Creator platform Eloelo has launched Dolphin AI, a new tool for making videos with artificial intelligence. The launch took place at FICCI…
Anthropic Launches AI Defense Program for Critical Infrastructure
Anthropic is launching a program that puts its most powerful AI models and engineers with companies protecting power grids and water utilities.
Chinese AI Firms Published Safety Tests for Just 3.6% of Models
A SemiAnalysis review found Chinese AI developers published model-specific safety tests for only 31 of 857 model releases.
Khabar 24h Editorial Desk — our explainers are prepared by the Khabar 24h editorial team using AI-assisted research tools, and every piece is reviewed by a human editor before publishing. We do not claim original reporting: our work is turning complex topics into simple, accurate summaries. Spotted an error? Write to contact@khabar24h.com — our corrections policy aims for same-day review.
More from this author →
Leave a Reply