Alibaba unveils 2.4T-parameter Qwen3.8 AI model

2026-07-20 · Source: Dataconomy · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Emerging Technologies & Innovation · Depth: Intermediate, quick

Summary

Alibaba has launched Qwen3.8, a large language model featuring 2.4 trillion parameters, which the company claims ranks "second only to Fable 5" in overall performance. This multimodal model significantly expands Alibaba's Qwen series, following Qwen3-Max (1 trillion parameters, September 2025) and Qwen3.5 (397 billion parameters, February 2026). A preview, Qwen3.8-Max-Preview, is accessible to developers through Alibaba's Token Plan, Qoder, and QoderWork, with open weights expected "soon." This marks a shift from Alibaba's previous strategy of keeping flagship Max models proprietary, though performance claims remain based on internal evaluations without independent verification.

Key takeaway

For AI Scientists evaluating new large language models, you should consider Qwen3.8's upcoming open-weight release as a potential new option for research and development. However, exercise caution regarding its performance claims, as they currently lack independent third-party verification. Prioritize testing the model's capabilities directly once open weights are available to assess its true competitive standing against established systems like Fable 5.

Key insights

Alibaba's 2.4T-parameter Qwen3.8 model is launching with a new open-weight strategy, despite unverified performance claims.

Principles

• Flagship models can transition to open-weight.
• Internal benchmarks require independent verification.
• Multimodality is a key LLM capability.

In practice

• Access Qwen3.8-Max-Preview via Token Plan.
• Monitor for Qwen3.8 open-weight release.
• Compare Qwen3.8 against Fable 5.

Topics

• Alibaba Qwen3.8
• Large Language Models
• Multimodal AI
• Open-Weight Models
• AI Benchmarking
• AI Competition

Best for: Research Scientist, AI Product Manager, AI Scientist, Director of AI/ML, Tech Journalist

Related on AIssential

• 😸 Alibaba’s 2.4T Qwen joins the AI race — Alibaba's 2.4T Qwen3.8-Max model, promised as open-weight, intensifies the competition between proprietary and accessible frontier AI.
• Alibaba's Qwen takes on Kimi K3 with open-weight Qwen 3.8, says model is "second only to Fable 5" — Alibaba's Qwen 3.8, a 2.4T parameter multimodal model, challenges Kimi K3 with open weights and claims near-Fable 5 performance.
• Alibaba's open Qwen 3.5 takes aim at GPT-5 mini and Claude Sonnet 4.5 at a fraction of the cost — Alibaba's Qwen 3.5 models achieve competitive performance against top Western LLMs at lower computational cost.
• Alibaba Says New AI Model Is Just Second to Anthropic’s Fable 5 — Alibaba's Qwen3.8 Max, with 2.4 trillion parameters, challenges top-tier AI models and will soon be open-weight.
• Something is afoot in the land of Qwen — Alibaba's Qwen team is experiencing significant upheaval, with lead researcher Junyang Lin and several other core members, including Binyuan Hui and Bowen Yu, announcing their resignations on March 4th, 2026.
• Alibaba just open-sourced a model that rivals GPT-5.2 — Alibaba's Qwen 3.5, with 17 billion active parameters, rivals top closed-source models while being open-weight.

See Counsel's argued verdicts on the open AI decisions leaders are weighing →

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Dataconomy.