Vercel's AI Gateway adds a free, 1M-token Ling 3.1 Flash
InclusionAI's hybrid reasoning model gets a context window nearly four times its previous limit, free to call through AI Gateway until pricing starts on 3 November.
Vercel added Ling 3.1 Flash 1019 from InclusionAI to AI Gateway on 30 October. It’s a newer checkpoint of the same 560-billion-parameter, 25-billion-active hybrid reasoning model Vercel already lists, but with a 1 million token context window, up from 262,144 on the existing listing. Vercel positions it for coding, multi-step reasoning and tool-using agents.
It’s callable as inclusionai/ling-3.1-flash-1019 through the AI SDK, the fx CLI, an OpenAI-compatible Chat Completions endpoint, or coding agents including Claude Code and Cursor:
import { streamText } from 'ai';
const result = streamText({
model: 'inclusionai/ling-3.1-flash-1019',
prompt: 'Summarise this codebase and flag the riskiest files.',
});It’s free to call until 3 November, after which Vercel bills it at the provider’s published rate, with no markup from AI Gateway itself on top.
Why it matters: a context window nearly four times the previous one is enough to hold a long document, a full small codebase or an extended agent run in a single call. The free window is short, so it’s worth testing against whatever long-context workload you already have before billing starts.
The caveat: the free period only runs five days, from 30 October to 3 November, so there isn’t much time to evaluate it properly before normal pricing applies.