GLM Flash Latest

The dedicated batch page for GLM-5.3-Flash — the current target of the Flash Latest alias — describes it as a native multimodal model from Z.ai purpose-built for efficient coding and long-horizon agent tasks, with a hybrid sparse and linear attention architecture that maintains accurate long-context behavior while redu Routing for the batch endpoint is single-provider via Together, with cache reads priced at $0.03 per 1M tokens, and the page exposes benchmarking data including GPQA Diamond and TAU-Benchvgi scores across hosting partners such as Phala (90.8% GPQA), BaseTen (84.2%), Wafer (88.3% GPQA, 80.0% TAU-Bench), DigitalOcean (89

OpenRouter's model page for the exact subject slug `z-ai/glm-flash-latest` documents that the alias "always redirects to the latest model in the GLM Flash family," and explicitly identifies the current redirect target as "Latest GLM 5.3 Flash." The page lists the slug as text-only with a 1.0M token context window and p Because `glm-flash-latest` is an OpenRouter redirect alias rather than a standalone checkpoint, the page itself does not publish weights, benchmarks, or pricing — those live on the sibling `z-ai/glm-5.3-flash` model page. What the page does establish for developers is that any code or routing using `z-ai/glm-flash-late