MiMo V2.6 models now available on AI Gateway
Xiaomi's MiMo V2.6 trio (Pro, Flash, Pro UltraSpeed) now available via Vercel's AI Gateway — multimodal models with 1M context and native agent support.
Why it matters
Practitioners building agents and multimodal apps gain three model options optimized for different workload profiles (reasoning vs. efficiency vs. speed) through a unified API platform. Expands the frontier-lab model menu beyond OpenAI/Anthropic/Google.
The key facts
13 to knowThree variants: MiMo V2.6 Pro (1.02T params, 42B active), Flash (309B params, 15B active), Pro UltraSpeed (20x output speed of Pro)
1M token context window; up to 128K output tokens
Native multimodal: text, image, audio, video understanding
Structured outputs and tool-use capabilities for agentic workflows
Available via Vercel AI Gateway with unified API, usage tracking, routing, failover, budgets
Integration: Vercel CLI for coding agents; model playground available
MiMo V2.6 Pro: 1.02T total parameters, 42B activated per token, sparse MoE architecture
MiMo V2.6 Flash: 309B total parameters, 15B activated per token, efficiency-optimized
MiMo V2.6 Pro UltraSpeed: 20x faster output than Pro, same capabilities
1M token context window, up to 128K output tokens
Native multimodal support: text, image, audio, video
Available via Vercel AI Gateway with unified API, cost tracking, routing, failover
Structured outputs and agent-native capabilities included
Go to the source
Vercel Blogvercel.com
Publisher excerpt: , , and from are now available on .MiMo V2.6 ProMiMo V2.6 FlashMiMo V2.6 Pro UltraSpeedXiaomiAI Gateway MiMo V2.6 combines coding, reasoning, and tool use with native text, image, audio, and video understanding. Its 1M token context supports long repositories, tool traces, and multi-session agent…