Xiaomi
Models published by Xiaomi, available through the AnyRouter API. Each can route across multiple upstream providers for availability and price.
MiMo V2 Flash is Xiaomi's open-weight MoE foundation model (309B total, 15B active) for high-speed inference, coding, and agent workflows. A hybrid global/sliding-window attention layout plus multi-token prediction makes generation 2.5-3.7x faster, and it ranks at the top of the open-source field on agent and code evaluations. AIHubMix serves it behind a free quota wire.
MiMo V2 Pro is Xiaomi's flagship foundation model for the agent era: over 1T total parameters with 42B active, a hybrid global/sliding-window attention layout widened to a 7:1 ratio from MiMo V2 Flash's 5:1, and a 1M-token context window. It targets high-intensity real-world agent work — long-horizon planning, workflow orchestration, and precise tool use — rather than chat alone. AIHubMix serves it behind a free quota wire.
MiMo V2.5 Pro is Xiaomi's flagship model for general agentic capabilities, complex software engineering, and long-horizon tasks. It independently completes professional tasks involving thousands of tool calls, with top rankings on SWE-bench Pro, ClawEval, and GDPVal. The 1M context window supports complete documents and complex agent workflows.
MiMo V2.5 is a native omnimodal model by Xiaomi delivering Pro-level agentic performance at roughly half the inference cost. It surpasses MiMo V2 Omni in multimodal perception across image and video understanding tasks, with a 1M context window ideal for agent frameworks where reasoning, rich perception, and cost efficiency all matter.
MiMo V2.6 Flash is Xiaomi's full-modality reasoning model for high-frequency invocations and large-scale professional workloads. It keeps the 1M-token context window and up to 128K output tokens of MiMo V2.6 Pro at roughly a third of the input price, making it the best-balanced pick when reasoning quality matters more than peak capability.
MiMo V2.6 Pro is Xiaomi's trillion-parameter, natively omni-modal flagship reasoning model for complex projects, long-horizon agentic tasks, coding, research, and professional knowledge work. It supports a 1M-token context window with up to 128K output tokens, and keeps the V2.5 API price while improving reasoning and long-task completion.