{"id":152065,"date":"2026-08-26T15:31:12","date_gmt":"2026-08-26T15:31:12","guid":{"rendered":"https:\/\/www.europesays.com\/ai\/152065\/"},"modified":"2026-08-26T15:31:12","modified_gmt":"2026-08-26T15:31:12","slug":"alibaba-targets-coding-and-office-tasks-with-low-cost-ai-mod","status":"publish","type":"post","link":"https:\/\/www.europesays.com\/ai\/152065\/","title":{"rendered":"Alibaba Targets Coding and Office Tasks With Low-Cost AI Mod\u2026"},"content":{"rendered":"

Alibaba<\/a> introduced Qwen3.8-Flash<\/a>, a new artificial intelligence model designed to deliver cost efficiency, the company said in a Wednesday (Aug. 26) post<\/a> on social platform X.<\/p>\n

\u26a1Meet Qwen3.8-Flash, a multimodal MoE and an early preview of the Qwen4 architecture, now open-weight!<\/p>\n

The production version Qwen3.8-Flash will be available soon via QwenCloud API at just $ 0.16\/1M input tokens and $ 0.47\/1M output tokens.<\/p>\n

125B parameters + 51B N-gram\u2026 pic.twitter.com\/SScnmzWS7O<\/a><\/p>\n

\u2014 Qwen (@Alibaba_Qwen) August 26, 2026<\/a><\/p>\n

Qwen3.8-Flash is a multimodal mixture of experts (MoE), and the production version will be available soon via the QwenCloud API at 16 cents per 1 million input tokens and 47 cents per 1 million output tokens. It features 125 billion parameters and 51 billion N-gram embeddings, with 6 billion activated per token, according to the post.<\/p>\n

\u201cDramatically lower training and inference costs: trained at just 1\/9 the cost of Qwen3.7-Plus, while outperforming it across the board with especially strong gains in coding and office tasks,\u201d Alibaba said in the post.<\/p>\n

\"Preferred<\/p>\n

We\u2019d love to be your preferred source for news<\/a>.<\/p>\n

Please add us to your preferred sources list so our news, data and interviews show up in your feed. Thanks!<\/p>\n

It was reported Aug. 15 that Alibaba\u2019s open-weight AI models<\/a> have become the world\u2019s top AI offerings. The company\u2019s models have seen more than 3 billion downloads worldwide over the last six months, surpassing those of Meta<\/a>, Google<\/a> and DeepSeek<\/a>.<\/p>\n

Alibaba said Qwen, its family of AI models, has open-sourced more than 460 models and its ecosystem has issued over 300,000 derivatives. Findings from open-source AI hub Hugging Face<\/a> showed that Google had 418 million downloads for the year, and Meta had 227 million.<\/p>\n

It was reported Aug. 14 that Apple<\/a> worked with Alibaba Group to develop and train an Apple AI model<\/a> specifically for the Chinese market.<\/p>\n

Apple has previously used third-party models to power the AI features it offers in the iPhones and other devices it sells in China. Now, following the collaboration with Alibaba, the company plans to launch its Apple Intelligence<\/a> suite of AI tools in China within months.<\/p>\n

Having its own model tailored for China would give Apple greater control over the AI experience it offers in the market and would make Apple the first foreign company to secure Chinese government approval to offer a proprietary model in the country.<\/p>\n

Alibaba reported Thursday (Aug. 20) that during the most recent quarter, its AI-related product<\/a> revenue saw triple-digit growth for the 12th consecutive quarter.<\/p>\n

\u201cWith our full-stack AI strategy, we have put Alibaba in a superior position to capture the substantial growth of demand for artificial intelligence and AI compute,\u201d Alibaba CEO Eddie Wu<\/a> said in an earnings press release.<\/p>\n

For all PYMNTS AI coverage, subscribe to the daily AI Newsletter<\/a>.<\/p>\n