Model catalogue
Xiaomi
MiMo-V2.5
xiaomi/mimo-v2.5
Specification
Published by the vendor; the platform bills at the target's declared price.
MiMo-V2.5
reasoning open weights toolsXiaomi xiaomi/mimo-v2.5
MiMo-V2.5 is a native omnimodal model by Xiaomi. It delivers Pro-level agentic performance at roughly half the inference cost, while surpassing MiMo-V2-Omni in multimodal perception across image and video understanding...
- Input $/1M
- $0.14
- Output $/1M
- $0.28
- Context
- 1.1M
- Max output
- 131K
- Released
- Apr 22, 2026
- Modalities
- textaudioimagevideo
Intelligence index-
Coding index57
Agentic index-
Cached input reads $0.003 per 1M · weights on Hugging Face as XiaomiMiMo/MiMo-V2.5
On this deployment
No routing target resolves to this model yet. Declare one and the router picks it up without retraining.
Not routable here yet
Add a target with metadata.model set to this id.
targets.yamlyaml
# targets.yaml - declare a target that runs MiMo-V2.5
targets:
- id: mimo-v2-5
kind: llm
name: MiMo-V2.5
metadata:
model: xiaomi/mimo-v2.5 # links the target to this catalogue entry
capabilities:
domains: [general]
supports_tools: true
context_window: 1050000
cost:
usd_per_1k_tokens: 0.000210Price among Xiaomi models
Input list price per million tokens; MiMo-V2.5 is the cheapest of 2 listed.
- 1$0.14MiMo-V2.5
- 2$0.43
More from Xiaomi
Other models by the same vendor in the catalogue.