MiMo V2.6 models now available on AI Gateway
By Steven Van ·
Xiaomi's MiMo V2.6 Pro, Flash, and Pro UltraSpeed join Vercel's AI Gateway with a 1M token context and up to 128K output tokens.
Vercel's AI Gateway now offers three MiMo V2.6 models from Xiaomi: Pro, Flash, and Pro UltraSpeed. All three combine coding, reasoning, and tool use with native text, image, audio, and video understanding, and share a 1M token context window with up to 128K output tokens and structured outputs.
- xiaomi/mimo-v2.6-pro: the larger sparse mixture-of-experts checkpoint, with 1.02T total parameters and 42B activated per token, for complex software engineering and long-running agent work. Priced at $0.43 input / $0.87 output per 1M tokens.
- xiaomi/mimo-v2.6-flash: 309B total parameters with 15B activated per token, the more efficient option for multimodal automation and everyday agent workflows. Priced at $0.14 input / $0.28 output per 1M tokens.
- xiaomi/mimo-v2.6-pro-ultraspeed: serves Pro at up to 20 times its output speed for interactive and latency-sensitive workflows, with the same capabilities. Priced at $4.35 input / $8.70 output per 1M tokens.
To use MiMo V2.6 in a coding agent, install the latest Vercel CLI and run the AI Gateway setup command, then select one of the three models. All three are also available to try in the model playground.
Vercel
The platform for frontend developers — deploy, preview, and scale web apps and AI agents with zero config.
View Vercel →