How to Install / Upgrade: Ling 3.0 Tiny is now available on AI Gateway

Ling 3.0 Tiny is now available on AI Gateway

Ling 3.0 Tiny from ANT Group is now on AI Gateway. It takes the free slot from Ling 3.0 Flash. The model is a MOE design with 7.9B total parameters and about 1.3B active per token, a 256K token context window, and up to 32K output tokens. It is built for responsive agents and instruction following.

To install or upgrade, switch your AI Gateway traffic from Ling 3.0 Flash to Ling 3.0 Tiny. Point your gateway model selection at Ling 3.0 Tiny for any free-tier or free-slot usage that previously used Flash. Keep the rest of your AI Gateway setup unchanged and deploy or restart the services that call the gateway so they pick up the new model target.

Watch the free window: use is free till 8:00am PT on 8/14, and Tiny now holds the free slot that Flash had. After you switch, verify that requests resolve to Ling 3.0 Tiny rather than Flash, that long prompts still fit within the 256K context window, and that responses stay within the 32K output limit.