DeepSeek V4.1 Flash brings native vision to AI Gateway

Vercel lists text and image input, a 1M-token context window, and access from Claude Code, Codex, and Cursor.

How this coverage works

This article combines reporting from 2 supporting Intel sources. It is updated as material evidence arrives; prior published revisions remain in the record.

DeepSeek has introduced V4.1 Flash with native visual understanding, and Vercel has made the model available through AI Gateway. The launch gives developers another model that accepts text and images together, including screenshots and charts.

Vercel lists a 1 million token context window and responses of up to 384,000 tokens, with reasoning, tool use, and prompt caching. Its announcement describes separate components for processing input and generating output. These are provider-published capabilities, rather than results from an independent evaluation.

Developers using AI Gateway can select deepseek/deepseek-v4.1-flash. Vercel also documents access from coding agents including Claude Code, Codex, and Cursor: install the latest Vercel CLI, run vercel ai-gateway setup, then select the model in the agent.

Update history1 updates

New facts extend one article instead of spawning duplicate write-ups across its source Intel pages.

Sep 10, 2026, 12:59 PM UTC
Revision 1 · initial
Initial launch report grounded in DeepSeek and Vercel primary announcements.