I Made Qwen 3.8 and Claude Opus 5 Build the Same Game

Qwen3.8 is one of the most interesting open-weight AI models for developers right now, so in this video I put Qwen 3.8-2.4T-A95B head-to-head with Claude Opus 5 to see how they actually compare on real coding work. Both models get the exact same prompt and are asked to build a complete playable 3D endless runner game in a single HTML file using Three.js. I compare Qwen 3.8 vs Claude Opus 5 while also looking at the Artificial Analysis Intelligence Index scores of 58 for Qwen and 63 for Claude Opus 5. 🔗 Relevant Links Qwen - https://qwen.ai/blog?id=qwen3.8 Hugging Face - https://huggingface.co/Qwen/Qwen3.8-2.4T-A95B ❤️ More about us Radically better observability stack: https://betterstack.com/ Written tutorials: https://betterstack.com/community/ Example projects: https://github.com/BetterStackHQ 📱 Socials Twitter: https://twitter.com/betterstackhq Instagram: https://www.instagram.com/betterstackhq/ TikTok: https://www.tiktok.com/@betterstack LinkedIn: https://www.linkedin.com/company/betterstack 📌 Chapters: 0:00 Qwen 3.8 vs Claude Opus 5 0:32 Qwen 3.8 2.4T Explained for Developers 1:25 Qwen 3.8 Tools and Hosted Inference 2:40 Getting Qwen 3.8 Setup and Hosted Inference 3:13 Qwen vs Claude: Real 3D Coding Challenge 5:30 Side by side Qwen vs Claude Comparison 6:06 Qwen 3.8 vs Claude Opus 5 Benchmarks 7:11 Speed, Verbosity and Open-Weight Access 7:51 Where Qwen 3.8 Is Actually Strong 8:30 Qwen 3.8 Weaknesses and Limitations 9:10 Should Developers Use Qwen 3.8? 9:22 Qwen 3.8 27B and Local AI Models 9:55 Qwen 3.8 vs Claude Opus 5: Final Verdict