Alibaba has released Qwen4-Max, its new flagship model. It is described as a 2.4-trillion-parameter mixture-of-experts system with built-in vision and a one-million-token context window. Alibaba claims it leads rival models on coding benchmarks. Unlike earlier Qwen releases, the weights for this model are not being published, although smaller Qwen4 models are planned.
Tencent, meanwhile, launched Hy4 Flash at $0.07 per million tokens. The company says it matches a leading small Western model on everyday tasks at a fraction of the cost. Benchmark and pricing comparisons here are the vendors’ own claims.
Why it matters for IT teams
Price competition in AI inference keeps intensifying, which is good news for high-volume workloads such as classification, summarisation and support automation. Before adopting any model, weigh data residency, compliance and support requirements alongside the headline price.




Leave a Reply