Ling 3.1 Flash is now available on AI Gateway
Vercel’s AI Gateway now offers Ling 3.1 Flash from InclusionAI, a hybrid reasoning model with a 262K-token context window, free until October 13, 2026.
Ling 3.1 Flash, developed by InclusionAI, has been added to Vercel’s AI Gateway, expanding the platform’s model catalog with a hybrid reasoning language model. The model features 560 billion total parameters and 25 billion active parameters per token, enabling efficient processing for complex tasks. Its 262K-token context window supports long documents, code, and extended task histories, making it suitable for coding, multi-step analysis, and tool-using agents.
The model is available for free through October 13, 2026, using the standard model ID. After the promotion ends, usage will begin billing automatically unless the free model ID is selected, which will stop serving instead of incurring charges. Users can access Ling 3.1 Flash via the AI Gateway model playground or integrate it into coding agents following the provided guide.
To deploy Ling 3.1 Flash, developers must select the model ID inclusionai/ling-3.1-flash when configuring their coding agents or API calls. The model is designed to handle workflows involving long documents and code, leveraging its large context window for extended task histories. Vercel’s AI Gateway simplifies model integration by offering a unified API for calling models, tracking usage, and managing routing and failover.
AI Gateway supports both Vercel-provided API keys and custom provider keys, allowing flexibility in how users interact with Ling 3.1 Flash and other models. The platform consolidates model access, cost tracking, and configuration under a single interface, streamlining deployment for developers. Ling 3.1 Flash joins AI Gateway’s catalog as a free-to-use option during the promotional period.