ToolsAugust 26, 2026via Vercel Blog
Qwen 3.8 Flash now available on AI Gateway
Why it matters
Qwen 3.8 Flash availability on Vercel's unified API layer signals continued model plurality in production stacks. Practitioners can now route agent and coding work through a vendor-neutral gateway with cost tracking and failover — making frontier-model choice a runtime decision, not an architecture lock-in.
Key signals
- Qwen 3.8 Flash now available via Vercel AI Gateway
- 1 million token context window; 65k token output limit
- Positioned for coding, tool use, multi-step agent work
- AI Gateway provides unified API, usage tracking, routing, failover, Zero Data Retention support
- No platform fee on inference; reflects provider pricing with no markup
- Integration ready with Claude Code, Codex, OpenCode, Cursor, Pi agents
- Published August 26, 2026
- Qwen 3.8 Flash available on Vercel AI Gateway
- 1 million token context window
- 65k token max output
- Works with Claude Code, Codex, OpenCode, Cursor, Pi agents
- Zero platform fee on inference; provider pricing reflected with no markup
- Integrates with AI Gateway's cost tracking, routing, failover, budgeting features
The hook
Alibaba's Qwen 3.8 Flash joins Vercel's AI Gateway — another frontier model gets commoditized into the dev-tools stack.
from Alibaba is now available on AI Gateway. It takes text and images as input, serves a context window of 1 million tokens, and can return up to 65k tokens in a response. Alibaba points it at coding, tool use, and multi-step agent work.Qwen3.8-Flash
To use Qwen3.8-Flash, set to in the :modelaliba…