MiMo-V2.5

Xiaomi's open-weights omnimodal workhorse (April 2026) and the #2 most-used model on OpenRouter by tokens. A 310B MoE (15B active) that takes text, image, video and audio input in one unified model with 1M context and no long-context surcharge, at some of the lowest prices of any frontier-adjacent model.

MiMo-V2.5 strengths

• #2 most-used on OpenRouter
• Native omnimodal input (text/image/video/audio)
• 1M context, no surcharge
• Open weights
• Ultra-low price

Pricing & context

Cost guide: a typical call of about 10K input + 2K output tokens costs roughly $0.002 at list prices. Worth modelling against cheaper tiers before committing high-volume traffic.

When to choose MiMo-V2.5

MiMo-V2.5 is best for high-volume agentic pipelines, long-document and multimodal processing where cost matters most. If your workload is more cost-sensitive, weigh it against gpt-oss-120b (≈$0.03 input /1M) first.

MiMo-V2.5 FAQ

How much does MiMo-V2.5 cost?

MiMo-V2.5 is priced at $0.105 per 1M input tokens and $0.28 per 1M output tokens (public API list price), with a 1M tokens context window. A typical call of about 10K input and 2K output tokens costs roughly $0.002.

What is MiMo-V2.5 best for?

MiMo-V2.5 by Xiaomi is best for high-volume agentic pipelines, long-document and multimodal processing where cost matters most.

How does MiMo-V2.5 pricing compare to Tencent Hy3?

MiMo-V2.5 input costs $0.105 per 1M tokens versus $0.063 for Tencent Hy3, roughly 1.7x more expensive on input. Output is $0.28 vs $0.21.

Is MiMo-V2.5 multimodal?

MiMo-V2.5 supports text, image, video, audio.