Alibaba
Available in Shortly AI
Qwen 3.8 Max
Released August 2, 2026
Qwen 3.8 Max is a 2.4-trillion-parameter MoE model delivering a comprehensive leap in coding and professional work. Autonomously codes and delivers complete projects spanning 10+ days. Handles hundreds of specialized tasks across legal, financial, design, and other professional domains, producing production-grade results end-to-end in a single conversation. Native visual understanding runs through the full cycle of planning, execution, and verification, enabling deep semantic analysis of ultra-long documents and extended video content. In long-horizon tasks, plans autonomously, iterates through closed feedback loops, and continuously evolves.
- Context window
- 1M tokens
- Maximum output
- 128K tokens
- Input pricing
- $2.00/M
- Output pricing
- $6.00/M
Independent benchmark
LiveBench performance
Benchmark data from LiveBench, release 2026-06-25. Scores are pinned and are not fetched on page load.
| Category | Score |
|---|---|
| Reasoning | 88.212 |
| Coding | 72.872 |
| Agentic Coding | 64.647 |
| Mathematics | 91.312 |
| Data Analysis | 78.413 |
| Language | 79.687 |
| Instruction Following | 74.083 |
Benchmarks measure selected evaluation tasks and do not guarantee performance for every prompt or workflow.
Public perspective
Reviews
Qwen 3.8 Max is an impressive but demanding model. It produces polished writing, follows detailed instructions, and handles complex tasks with unusual depth. Longer tests also show that it can complete substantial work in one pass, although it often takes time to get there. The writing quality is a particular strength, especially for creative and non-technical work.
The experience is less consistent when Qwen 3.8 Max is used as an autonomous coding agent. It can plan well, reason through a large task, and sometimes deliver an excellent result, but it can also lose track of files, repeat itself, or struggle to use tools reliably. There are also reports of unfinished or broken implementations being treated as complete, so important work should still be reviewed by a person.
The main trade-off is speed and cost. Qwen 3.8 Max is worth considering when quality matters more than response time, especially for careful research, creative writing, and high-effort coding. For quick everyday questions or heavily automated workflows, a faster and cheaper model may be easier to use. Its token usage can also be high in practice.
Research reviewed August 26, 2026.
FAQ
Similar AI models
Available with one subscription
Start with Qwen 3.8 Max in Shortly AI
Keep your conversations, files and model choices together in one workspace.