Which Chinese LLM API should you use? One guide per workload — every pick backed by the live price table, with monthly cost estimates and the same rules as the picker on the models page.
Coding workloads split three ways: hard reasoning on complex refactors, long agentic sessions across a repo, and high-volume generation of boilerplate. Here is the pick for each — …
Agent workloads are context-hungry: long histories, tool loops, whole documents. The winning move is a model with a 256K context window and cached-input billing so repeated context…
For support bots, copilots and general assistants, the price of quality is everything: the conversation layer runs at high volume and every cent-per-token compounds. These are the …
Reasoning workloads reward capability first, but the gap between "best" and "second best" rarely justifies a 10× price jump. The Chinese flagship tier is unusually competitive righ…
RAG has two cost centers: the embedding model (cheap but multiplied by your corpus size) and the generation model (your only real lever). The Bridge stack below keeps both at the b…
Only one model in the Bridge catalog accepts images today — and it happens to be the strongest Chinese-language model in the group. That makes this guide short and definitive.…
At 50M+ tokens a month, model choice is an infrastructure decision. The high-volume tier is where Chinese open-weight economics and discounted channels do their best work — these a…