#149149th of 168 in ML frameworks and libraries
mmf
PyTorch framework for vision and language multimodal research
Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 5.6k
- License
- custom license
- Last commit
- Oct 2026
- Last release
- Aug 2019
- Language
- Python
5,604 on 21 Apr 20265,633 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
MMF is a PyTorch codebase from Facebook AI Research with reference implementations of vision and language models and support for distributed training. It also serves as a starter codebase for challenges such as Hateful Memes, TextVQA, TextCaps and VQA. It was formerly called Pythia.
Who it is for: Researchers building vision and language models in PyTorch
Strengths
- Reference implementations of vision and language models from FAIR research projects
- Distributed training supported on PyTorch
- Starter code for Hateful Memes, TextVQA, TextCaps and VQA challenges
- Documentation site with feature list and video overview
Weaknesses
- Last tagged release is from 2019-08-26
- Research codebase, not a packaged app or inference server
- License detected as NOASSERTION; README only says BSD
- README gives no hardware requirements or model list inline
What it needs
- Needs PyTorch
Quick answers
Is mmf open source?
Its code is public on GitHub under its own license terms; read them before you use it at work.
Also in ML frameworks and libraries
See all 168| Rank | Project | Score |
|---|---|---|
| 1 | transformersPython library defining open models for text, vision, audio and multimodal tasks | 94 out of 100 |
| 2 | pytorchPython tensor library with GPU support and tape-based autograd | 93 out of 100 |
| 3 | ultralyticsPython library and CLI for training and running YOLO vision models | 93 out of 100 |
| #4 | pytorch-image-modelsPyTorch library of image models, pretrained weights and training scripts | 93 out of 100 |
| #5 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 92 out of 100 |