DeepSeek

For months the advice was simple: want capable AI on a budget? Use DeepSeek. The official V4 Pro launch just ended that. With OpenAI Luna now at $0.20 per million input tokens, the cheapest-model crown is split three ways.

DeepSeek has warned developers of a significant API price increase. Its V4 Flash is the cheapest capable model on the market with no real rival on price-performance — but OpenAI just slashed GPT-5.6 Luna by 80 %, and that changes everything.

Within a five-week window, the AI industry shipped GPT-5.6, Kimi K3, Gemini 3.6 Flash, and GLM-5.2. But the pipeline is far from empty. Here is what we know about the next wave: GLM 5.5, Alibaba Qwen 3.8 Max with confirmed open weights, DeepSeek V4 full release, and the next Google Gemini flagship — all likely to land before September.