Tencent Hy4 Flash has cut its base API rates below those of competitors, while offering a context window of one million tokens. Put simply, a context window is how much text a model can read at once; one million tokens means you can fit a very long document in and then ask questions about it. For developers processing long texts or making API calls in bulk, price and context-window size directly affect costs. As for its claim that its office-work capabilities are close to Claude Haiku’s, The Defiant reported that the conclusion came from a blind test organized by Tencent itself, making it the company’s own assessment rather than an independent verification. In the short term, don’t focus on a single headline; trading volume and the price range are more telling.