Developers wanting a fast, MIT-licensed open model can download it free; hosted API prices could not be read.
Kenning's take
What it does
MiMo-V2-Flash is a large language model from Xiaomi with 309 billion total and 15 billion active parameters. It mixes sliding-window and global attention to cut memory use and uses multi-token prediction to speed up output, with a context window up to 256K tokens. The weights are on Hugging Face under the MIT licence, and it can be used in Xiaomi MiMo Studio and the MiMo API platform.
- 309B total, 15B active parameters
- Up to 256K token context
- Hybrid sliding-window and global attention
- Multi-token prediction for faster output
- MIT-licensed weights, including MTP layers
Who it suits. Developers who want a fast open model for coding and agent tasks they can self-host.
Key facts
- Category
- AI models
- Made by
- Xiaomi
- Based in
- China
- How to get it
- free weights on Hugging Face; hosted in Xiaomi MiMo Studio and through the MiMo API platform
- Works on
- self-hosted, Web, API
- Free plan
- Free MIT-licensed weights
FAQ
Is MiMo-V2-Flash open source?
Its weights are on Hugging Face under the MIT licence.
How big is MiMo-V2-Flash?
309 billion total parameters, with 15 billion active per token.
How long is its context window?
Up to 256K tokens.
Facts from the maker's own pages and the sources listed. Checked 2026-10-11.
Similar tools
ACE-Step 1.5
Open-source music generation model that writes full songs with lyrics and runs on ordinary consumer GPUs.
AlphaGenome Atlas
Google DeepMind database of AI-predicted effects for every possible single-letter change in human DNA.
Amazon Nova 2
Amazon's second family of Nova AI models, covering reasoning, speech conversation and combined text-and-image generation.
Cisco Antares
Cisco's small open-weight models that find where known security vulnerabilities sit in a codebase, running locally.