Vercel adds Ling 3.0 Flash to AI Gateway
Vercel announced the integration of inclusionAI's Ling 3.0 Flash (inclusionai/ling-3.0-flash-free) into Vercel AI Gateway. The Mixture-of-Experts model supports thinking and non-thinking execution modes for agentic workflows, with free access through August 3.
Expanding AI Gateway with agent-focused MoE models gives developers more flexible options for building high-performance autonomous AI workflows.
- –Features dual operational modes (thinking and non-thinking) to balance latency against deep reasoning needs.
- –Leverages a Mixture-of-Experts (MoE) architecture tailored for agentic tasks.
- –Free promotional window through August 3 encourages rapid developer experimentation across Vercel applications.
DISCOVERED
3h ago
2026-07-23
PUBLISHED
3h ago
2026-07-23
RELEVANCE
AUTHOR
vercel_dev