Uploaded August 2026 | Updated September 2026, 2 weeks ago
🚀 The LangChain 10 Days FREE Bootcamp is live: 10 lessons, free AI models only, from your first API call to a production grade RAG agent. Start with Day 0 for the roadmap and setup.
📺 Full playlist: youtube.com/watch?v=KJ3_NExk7-Q&list=PLW4pPr9JCovI&index=1
----------
Qwen 3.8 Max vs Claude Opus 5, same prompt, same time, and I tracked the latency, the tokens and the exact dollar cost of both.
Alibaba just open sourced the Qwen 3.8 Max weights, so I gave both models one job: build a 3D model of the 1903 Wright Flyer in a single HTML file using Three.js with orbit controls. Qwen returned a plane I could orbit, zoom and throttle. Opus returned a page where I could not see anything at all, and it took 83% of the $1.67 I spent. This is one prompt and one run, not a benchmark suite, so treat it as a data point and not a verdict on every task.
⏱ Chapters:
0:00 What I tested and what happened
0:45 The cost split before we start
1:19 The Opus 5 result: a blank screen
2:03 Qwen 3.8 Max: 2.4T parameters, 95B active
2:59 1M context, long coding runs and benchmarks
4:04 The prompt: 1903 Wright Flyer in Three.js
5:06 Side by side pricing on OpenRouter
6:21 Latency and who finished first
6:58 Saving qwen.html and opus.html
7:50 Opening the Opus 5 build
8:01 Opening the Qwen 3.8 Max build
8:32 Orbit controls, throttle and a floating bug
9:50 Final token use and cost split
11:25 Verdict
🔗 The exact prompt I used:
Build a 3D model of a 1903 Wright Flyer in a single HTML file using Three.js with orbit controls.
🔗 Compare model pricing and latency: openrouter.ai/models
🔗 Three.js docs: threejs.org
📺 Watch next:
Nemotron 3.5 Lightning vs Muse Glimmer on RTX 5090: Full Benchmark - youtube.com/watch?v=n74N6p5so7o
Gemini 3.5 Flash vs Qwen 3.7 - Qwen 3.7 Beat It Twice - youtube.com/watch?v=VaiGQQpL48M
Gemini 3.5 Flash vs Claude Sonnet 4.6 vs Opus 4.7 - Speed, Cost & Quality Tested - youtube.com/watch?v=Pi1BbG48w1I
🎓 Go deeper - my Udemy course:
Master Langchain v1 and Ollama - Chatbot, RAG and AI Agents: kgptalkie.com/langchain
If this saved you a few dollars in API credits, hit like. Tell me in the comments which prompt you want me to run next on these two models, and subscribe with the bell so you catch the next model test.
#Qwen #ClaudeOpus5 #AICoding
🚀 The LangChain 10 Days FREE Bootcamp is live: 10 lessons, free AI models only, from your first API call to a production grade RAG agent. Start with Day 0 for the roadmap and setup.
📺 Full playlist: youtube.com/watch?v=KJ3_NExk7-Q&list=PLW4pPr9JCovI&index=1
----------
Qwen 3.8 Max vs Claude Opus 5, same prompt, same time, and I tracked the latency, the tokens and the exact dollar cost of both.
Alibaba just open sourced the Qwen 3.8 Max weights, so I gave both models one job: build a 3D model of the 1903 Wright Flyer in a single HTML file using Three.js with orbit controls. Qwen returned a plane I could orbit, zoom and throttle. Opus returned a page where I could not see anything at all, and it took 83% of the $1.67 I spent. This is one prompt and one run, not a benchmark suite, so treat it as a data point and not a verdict on every task.
⏱ Chapters:
0:00 What I tested and what happened
0:45 The cost split before we start
1:19 The Opus 5 result: a blank screen
2:03 Qwen 3.8 Max: 2.4T parameters, 95B active
2:59 1M context, long coding runs and benchmarks
4:04 The prompt: 1903 Wright Flyer in Three.js
5:06 Side by side pricing on OpenRouter
6:21 Latency and who finished first
6:58 Saving qwen.html and opus.html
7:50 Opening the Opus 5 build
8:01 Opening the Qwen 3.8 Max build
8:32 Orbit controls, throttle and a floating bug
9:50 Final token use and cost split
11:25 Verdict
🔗 The exact prompt I used:
Build a 3D model of a 1903 Wright Flyer in a single HTML file using Three.js with orbit controls.
🔗 Compare model pricing and latency: openrouter.ai/models
🔗 Three.js docs: threejs.org
📺 Watch next:
Nemotron 3.5 Lightning vs Muse Glimmer on RTX 5090: Full Benchmark - youtube.com/watch?v=n74N6p5so7o
Gemini 3.5 Flash vs Qwen 3.7 - Qwen 3.7 Beat It Twice - youtube.com/watch?v=VaiGQQpL48M
Gemini 3.5 Flash vs Claude Sonnet 4.6 vs Opus 4.7 - Speed, Cost & Quality Tested - youtube.com/watch?v=Pi1BbG48w1I
🎓 Go deeper - my Udemy course:
Master Langchain v1 and Ollama - Chatbot, RAG and AI Agents: kgptalkie.com/langchain
If this saved you a few dollars in API credits, hit like. Tell me in the comments which prompt you want me to run next on these two models, and subscribe with the bell so you catch the next model test.
#Qwen #ClaudeOpus5 #AICoding
