Merge Gateway announces 90% off GLM 5.3 Flash through September
On the model gateway, input drops to $0.012 and output to $0.04 per million tokens. The company argues this puts cost per task far below rival models.
Model gateway service Merge announced a 90% discount on GLM 5.3 Flash running through the end of September. The prices it gives are $0.012 per million input tokens, $0.04 output and $0.003 cached.
For comparison: on Z.ai's own list, GLM 5.3 Flash carries a launch discount of $0.075 per million input tokens and $0.25 output; the undiscounted rates are $0.15 and $0.50.
In its announcement Merge says the model scores 57 out of 100 on the Artificial Analysis intelligence index, and that among models scoring above 55 its gateway brings cost per task down to $0.009. In the chart the company shared, other models meeting that threshold range from $0.40 to $0.53 per task. That comparison is Merge's own presentation; it has not been independently verified.
Merge Gateway launched in March as a service giving access to models from multiple providers through a single endpoint, with routing and cost controls. The gateway normally adds 5% on top of model cost.
GLM 5.3 Flash is the multimodal model Z.ai announced on 26 August, with 320 billion parameters of which 18 billion are activated per token.
For details see Merge Gateway and Z.ai's pricing page.
