[AINews] Qwen 3.8 Max(2.4T) and 27B, new open weights models for Coding and Cowork
Quick Answer
Alibaba's Qwen 3.8 Max, a 2.4T parameter model, is set to release open weights next week, showcasing capabilities in autonomous coding, chip design, and multimodal reasoning.
Quick Take
It achieved top 13% performance in a competitive data science challenge, with API pricing at $2/M input and $6/M output tokens.
Key Points
- Qwen 3.8 Max features 2.4 trillion parameters, focusing on coding and multimodal tasks.
- Achieved top 13% in WWW2025 Multimodal Dialogue Intent Recognition Challenge.
- Autonomous coding for over 10 days and chip design optimization with 500+ iterations.
- API pricing set at $2/M input tokens and $6/M output tokens.
- Open weights for both Qwen 3.8 Max and Qwen 3.8-27B to be released next week.
DeepSignal Analysis
What happened
Alibaba is set to release open weights for its Qwen 3.8 Max model, which has 2.4 trillion parameters, next week. The model has demonstrated strong performance in various tasks, including coding and chip design, and will be available via API at specified pricing. Additionally, Qwen 3.8-27B will also have open weights released.
Key evidence
- Alibaba announced that Qwen3.8-Max will have open weights available next week, alongside Qwen3.8-27B.
- The API pricing for Qwen models is set at $2 per million input tokens and $6 per million output tokens.
- Qwen3.8-Max achieved a ranking of #4 in the Frontend Code Arena with an Elo score of 1,668.
Why it matters
The release of Qwen 3.8 Max signifies a notable advancement in open-weight models, potentially positioning Alibaba as a strong competitor against established Western models. Its capabilities in autonomous coding and multimodal reasoning may influence the development of AI applications across various industries. The competitive performance metrics suggest that open models are gaining traction in areas traditionally dominated by proprietary systems.
What to watch
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from Latent Space
See more →
Ontologies Are So Back: Why AI Agents Are Reviving the Semantic Web
Frank Coyle at AIEWF 2026 emphasized the revival of ontologies as essential 'logical guardrails' for effective AI agents, integrating them with for better reasoning. Neo4j's Emil Eifrem highlighted three ontology types to enhance agent scalability, while Kingsley Idehen discussed the challenges and benefits of maintaining ontologies in AI systems.
![[AINews] SpaceXAI launches Grok 4.5, first Opus-class model post Cursor acquisition](https://substackcdn.com/image/fetch/$s_!8D6O!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fpbs.substack.com%2Fmedia%2FHMuQw2BXUAAJaQd.png)
