Alibaba Qwen Releases Qwen3.8-Max: A 2.4 Trillion Parameter MoE Model and the Most Capable in the Qwen Family

Alibaba's Qwen team has made Qwen3.8-Max broadly available and confirmed its open weights will ship next week, along with a second checkpoint, Qwen3.8-27B. Qwen3.8-Max is a 2.4-trillion-parameter mixture-of-experts model that accepts text, image, and video input as returns text. The hosted API is deployable for companies of any size and is compatible with OpenAI and DashScope, while the open weights are a multi-node datacenter artifact. Alibaba has not disclosed the activated-parameter count, making serving cost unknown. Qwen3.8-27B is the checkpoint that fits ordinary on-premise GPU hardware. The published feature set maps onto four industries: software engineering, legal and financial document review, media and e-commerce operations, and design. The model page lists a 1M-token context window, with maximum input of 991K tokens (983K with thinking enabled) and maximum output of 131K tokens. Pricing is $2.00 per 1M input tokens and $6.00 per 1M output tokens, with cached input costing $0.25 per 1M tokens. Supported capabilities include function calling, structured outputs, batches, prefix completion, and fine-tuning, along with five built-in tools on the Responses API. In benchmarks, Qwen3.8-Max scores 86.6 on Terminal-Bench 2.1, behind GPT-5.6 Sol (max) at 88.8 but ahead of Claude Opus 4.8 and Claude Fable 5 at 84.6. It leads on PaperBench and IFBench, and shows gains in multimodal and agentic tasks over its predecessor, though the multimodal table benchmarks against Qwen3.7-Plus, not Qwen3.7-Max. The model's RL scaling curve peaks at 0.725 near 4,000 training environments, then declines.