I think the next 6 month will be a golden age for inference. The weights will be available for anyone to download from hugging face which could create a price fall once providers start competing for your business.

The pricing is 40% of Kimi K3 at the same Model size. It will be interesting to see benchmarks tomorrow and to see how the pricing evolves across this size class.

Blog post:

https://qwen.ai/blog?id=qwen3.8 (https://news.ycombinator.com/item?id=49150470)

We're reaching a point where API cost is no longer the bottleneck for agentic workflows