Alibaba launches Qwen-3 Max, its most advanced AI model yet, designed to compete with giants like OpenAI’s ChatGPT and Google’s Gemini. This powerful large language model (LLM) boasts over one trillion parameters, offering top-tier performance in reasoning, coding, and multilingual tasks. Available for free via the Qwen app or website, Qwen-3 Max is already making waves, ranking third on LMArena’s text leaderboard. It outshines standard GPT-5 and delivers high accuracy with fewer errors, setting a new benchmark for AI innovation.
Why Qwen-3 Max Stands Out?
Qwen-3 Max excels in multiple areas, from solving complex coding problems to understanding lengthy documents. Its advanced features make it a strong rival to leading AI models. Here’s what you need to know:
- Massive Scale: Pretrained on 36 trillion tokens with a context length of one million tokens.
- Top Performance: Scores 69.6 on SWE-Bench for coding and 74.8 on Tau2-Bench for tool-calling proficiency.
- Enhanced Accuracy: Improves reasoning, math, logic, and science tasks with reduced hallucinations.
- Multilingual Support: Handles English and Chinese instructions reliably.
- Free Access: Available on the Qwen app or website for all users.
Also Read: Key Learnings From The Failure Percentage Of Startups?
Alibaba also teases Qwen-3 Max Thinking, a model in training that shows promise, scoring 100% on benchmarks like AIME 25 and HMMT. While it trails behind Gemini 2.5 Pro and Claude Opus 4.1 in some rankings, Qwen-3 Max surpasses DeepSeek V3.1 in key tests, proving its competitive edge.
To start using Qwen-3 Max, open the Qwen app on iOS or Android, where it’s set as the default model. Alternatively, visit the Qwen website, select the model from the top left corner if needed, and begin chatting. Explore its capabilities and see how it stacks up against your favorite AI chatbot today!
More News To Read: Meta Boosts Sports Engagement In Its Apps