language	library_name	license_name
en	mir	MPL-2.0 + Commons Clause 1.0

MIR (Machine Intelligence Resource)
_{A naming schema for AIGC/ML work.}

The MIR classification format seeks to standardize and complete a hyperlinked network of model information, improving accessibility and reproducibility across the AI community.

This repo is an example development implementation of autogenerated model inference parameters and code with the MIR schema. This is the sister repo to our 🤗HuggingFace MIR project which is an archive of model state dict layer information that also uses the MIR schema.

MIR Schema:

Note

Example:

mir : // model . vit . clip-l : stable-diffusion-xl

mir : // model .    lora      .    hyper    :   flux-1
  ↑      ↑         ↑               ↑            ↑
 [URI]:[Domain].[Architecture].[Series]:[Compatibility]

DOMAIN

↑Most Specific/Decentralized

Dev

Pre-release or under evaluation items without an identifier in an expected format Anything in in-training, pre-public release, and items under evaluation

Meant to be created by anyone, derived from code and file analysis

• Contextual
• Layers of neural networks
• Dynamic

Model

Publicly released machine learning models with an identifier in the database Model weight tensors with arbitrary locations and quantitative dimensions

Meant to be created by file hosts, derived from research pre-prints

• Contextual
• Layers of neural networks
• Fixed

Ops

References to specific optimization or manipulation techniques Algorithms, optimizations and procedures for models

Meant to be created by code libraries, derived from research pre-prints

• Universal
• Attributes of neural networks
• Dynamic

Info

Metadata of layer names or settings with an identifier in the database Information about the model and tensor specifications

Meant to be created by standards community, derived from code and file analysis

• Universal
• Attributes of neural networks
• Fixed

↓Most General/Centralized

ARCHITECTURE

Abbreviation	Description
AET	Autoencoding Transformer
ART	Autoregressive Transformer
BRNN	Bi-directional Recurrent Neural Network
CNN	Convolutional Neural Network
CONTROLNET	Controlnet
DETR	Detection Transformer
GAN	Generative Adversarial Model
GRU	Gated recurrent unit
LORA	Low-Rank Adaptation
LSTM	Long Short-Term Memory
MOE	Mixture of Experts
RBM	Restricted Boltzmann machine
RCNN	Region-based Convolutional Neural Network
RESNET	Residual Network
RNN	Recurrent Neural Network
SSM	State-Space Model
STST	Sequence-to-Sequence Transformer
TAE	Tiny Autoencoder
UNCLASSIFIED	Unknown
VAE	Variable Autoencoder
VLA	Vision Language Action
VIT	Vision Transformer

--

SERIES

Rules

• Lowercase, hyphen only
• Remove parameter size, non-breaking semantic versioning, library names

Example: tencent-hunyuan/hunyuandiT-v1.2-diffusers
SERIES : hunyuandit-v1

Example: black-forest-labs/FLUX.1-dev
SERIES : flux1-dev

In regex (roughly)

BREAKING*SUFFIX = r".*(?:-)(prior)$|.*(?:-)(diffusers)$|.\*[\*-](\d{3,4}px|-T2V$|-I2V$)"
PARAMETERS*SUFFIX = r"(\d{1,4}[KkMmBb]|[.*-]\d+[\._-]\d+[Bb][._-]).\*?$"
SEARCH*SUFFIX = r"\d+[.*-]?\d+[BbMmKk](it)?|[._-]\d+[BbMmKk](it)?"

COMPATIBILITY

Rules

An additional SERIES label for identifying cross-compatibility

---

If you would like to regenerate or update the example file here, use nnll:

MIR is inspired by:

• AIR-URN project by CivitAI
• Spandrel super-resolution registry by chaiNNer
• SDWebUI Model Toolkit by silveroxides

About this project

````