Qwen 3.8 Flash now available on AI Gateway
Vercel’s AI Gateway now supports Alibaba’s Qwen 3.8 Flash model, enabling text and image input with a 1 million token context window and 65k token output. The model targets coding, tool use, and multi-step agent workflows.
Vercel has integrated Alibaba’s Qwen 3.8 Flash into its AI Gateway, expanding the platform’s model offerings. The model accepts both text and images as input, supports a context window of up to 1 million tokens, and can generate responses of up to 65,000 tokens. Alibaba positions the model for coding tasks, tool integration, and multi-step agent workflows, aligning with developer needs for high-capacity processing.
To deploy Qwen 3.8 Flash in coding agents, developers must follow Vercel’s coding agents guide and execute the command vercel ai-gateway coding-agents setup. This process enables connections to agents such as Claude Code, Codex, OpenCode, Cursor, and Pi, with the model accessible under the identifier alibaba/qwen3.8-flash. The setup streamlines integration while maintaining compatibility with existing agent frameworks.
AI Gateway provides a unified API for model interactions, including usage tracking, cost management, and configurable retries, failover, and performance optimizations to enhance uptime beyond provider guarantees. The service includes built-in custom reporting, Zero Data Retention support, budget controls for API keys, and routing rules to manage traffic and costs efficiently.
Pricing for Qwen 3.8 Flash on AI Gateway reflects the provider’s rates without additional markup, and the platform does not charge a fee for inference, even for Bring Your Own Key (BYOK) requests. This pricing structure aims to reduce costs for developers while ensuring transparent billing aligned with Alibaba’s published rates.