Alibaba releases AI model Qwen to challenge US rival Anthropic’s Fable

Alibaba Group Holding has released its biggest-ever AI model, claiming performance on a par with global leader Anthropic in the latest Chinese breakthrough to challenge US rivals.

The new Qwen3.8-Max is built on 2.4 trillion parameters and ranks higher on several benchmarks than the headline-grabbing Kimi K3 from Moonshot that was recently unveiled.

Alibaba shared results showing it delivered comparable or sometimes better scores than Anthropic’s Fable 5, a cutting-edge artificial intelligence model that was temporarily put under export controls by the US due to its advanced capabilities.

The debut comes days after Moonshot’s Kimi sent ripples through stock markets and Silicon Valley as it showed that Chinese developers are quickly catching up with the top models crafted by Anthropic and OpenAI, despite relatively constrained computing resources. DeepSeek also just expanded access to its latest model, V4 Flash, while ByteDance and MiniMax Group unveiled new video generators on July 31.

“Many investors continue to underestimate Chinese AI models because of US chip restrictions or general scepticism,” said Vey-Sern Ling, managing director at Union Bancaire Privee.

“In reality, the gap is probably much closer and narrowing fast. Alibaba’s Qwen 3.8 is another proof point, following Kimi K3.”

Alibaba’s shares surged as much as 7.3 per cent, the most in nearly a month, in Hong Kong on Aug 3.

The Hangzhou-headquartered internet pioneer will release the Qwen3.8-Max weights for public download next week, which will allow users to customise the technology, marking the next major move in the intensifying race among China’s AI contenders that include Moonshot, DeepSeek, Z.ai and ByteDance.

While DeepSeek’s latest is by far the most affordable among new marquee releases, Alibaba’s Qwen offering is also priced aggressively at US$2 (S$2.60) per million input tokens and US$6 per million outputs. Each AI system will use a different number of tokens to handle tasks, but that still makes Alibaba’s model look attractive compared with the best from the US leaders.

The new Qwen3.8-Max performs well in autonomous coding and long-horizon execution and was able to independently perform a software engineering project over 16 days in internal testing, Alibaba said.

The new system has also improved in efficiency, activating only some parts when in use to reduce computational costs and latency. BLOOMBERG