README.md

---
pipeline_tag: image-classification
license: apache-2.0
base_model: timm/maxvit_tiny_tf_512.in1k
library_name: kerasformers
tags:
- keras
- kerasformers
- image-classification
- maxvit
- backbone
- arxiv:2204.01697
- pytorch
- jax
- tf
---

## ***See [our collection](https://huggingface.co/collections/kerasformers/maxvit-6a6bd734b6d773748325c03c) for all versions of MaxViT.***

# Run MaxViT with Keras 3: JAX, PyTorch, or TensorFlow

[![GitHub](https://img.shields.io/badge/GitHub-KerasFormers-black?logo=github)](https://github.com/IMvision12/KerasFormers) [![Docs](https://img.shields.io/badge/Docs-MaxViT-blue)](https://imvision12.github.io/KerasFormers/classification_backbones/) [![Collection](https://img.shields.io/badge/HF-MaxViT%20collection-yellow)](https://huggingface.co/collections/kerasformers/maxvit-6a6bd734b6d773748325c03c)

# kerasformers/maxvit_tiny_tf_512_in1k

Paper: [MaxViT: Multi-Axis Vision Transformer (arXiv:2204.01697)](https://arxiv.org/abs/2204.01697) · [HF Papers](https://huggingface.co/papers/2204.01697)

MaxViT combines blocked local and dilated global attention (multi-axis) in a hierarchical CNN/Transformer hybrid.

For more details on the model, please go to the upstream [model card](https://huggingface.co/timm/maxvit_tiny_tf_512.in1k).

Pure-**Keras 3** conversion of [`timm/maxvit_tiny_tf_512.in1k`](https://huggingface.co/timm/maxvit_tiny_tf_512.in1k) for [kerasformers](https://github.com/IMvision12/KerasFormers). One implementation runs unmodified on **TensorFlow / Torch / JAX**.

This is an **image-classification / backbone** checkpoint (`MaxViTImageClassify` / `MaxViTModel`).

## ✨ Quick start

```python
import os
os.environ["KERAS_BACKEND"] = "torch"  # or "jax" / "tensorflow"

from PIL import Image
import numpy as np
from kerasformers.models.maxvit import MaxViTImageClassify, MaxViTModel

model = MaxViTImageClassify.from_weights("kerasformers/maxvit_tiny_tf_512_in1k")
backbone = MaxViTModel.from_weights(
    "kerasformers/maxvit_tiny_tf_512_in1k", as_backbone=True
)

image = Image.open("your_image.jpg").convert("RGB")
image = image.resize((224, 224))
x = np.asarray(image, dtype="float32")[None]  # (1, H, W, 3)
print(model(x).shape)  # (1, num_classes)
feats = backbone(x)
print(len(feats), [tuple(f.shape) for f in feats])
```

Load any MaxViT variant the same way with `from_weights("kerasformers/<variant>")`:

| Variant | Hub |
|---|---|
| `maxvit_base_tf_224_in1k` | [`kerasformers/maxvit_base_tf_224_in1k`](https://huggingface.co/kerasformers/maxvit_base_tf_224_in1k) |
| `maxvit_base_tf_224_in21k` | [`kerasformers/maxvit_base_tf_224_in21k`](https://huggingface.co/kerasformers/maxvit_base_tf_224_in21k) |
| `maxvit_base_tf_384_in1k` | [`kerasformers/maxvit_base_tf_384_in1k`](https://huggingface.co/kerasformers/maxvit_base_tf_384_in1k) |
| `maxvit_base_tf_384_in21k_ft_in1k` | [`kerasformers/maxvit_base_tf_384_in21k_ft_in1k`](https://huggingface.co/kerasformers/maxvit_base_tf_384_in21k_ft_in1k) |
| `maxvit_base_tf_512_in1k` | [`kerasformers/maxvit_base_tf_512_in1k`](https://huggingface.co/kerasformers/maxvit_base_tf_512_in1k) |
| `maxvit_base_tf_512_in21k_ft_in1k` | [`kerasformers/maxvit_base_tf_512_in21k_ft_in1k`](https://huggingface.co/kerasformers/maxvit_base_tf_512_in21k_ft_in1k) |
| `maxvit_large_tf_224_in1k` | [`kerasformers/maxvit_large_tf_224_in1k`](https://huggingface.co/kerasformers/maxvit_large_tf_224_in1k) |
| `maxvit_large_tf_224_in21k` | [`kerasformers/maxvit_large_tf_224_in21k`](https://huggingface.co/kerasformers/maxvit_large_tf_224_in21k) |
| `maxvit_large_tf_384_in1k` | [`kerasformers/maxvit_large_tf_384_in1k`](https://huggingface.co/kerasformers/maxvit_large_tf_384_in1k) |
| `maxvit_large_tf_384_in21k_ft_in1k` | [`kerasformers/maxvit_large_tf_384_in21k_ft_in1k`](https://huggingface.co/kerasformers/maxvit_large_tf_384_in21k_ft_in1k) |
| `maxvit_large_tf_512_in1k` | [`kerasformers/maxvit_large_tf_512_in1k`](https://huggingface.co/kerasformers/maxvit_large_tf_512_in1k) |
| `maxvit_large_tf_512_in21k_ft_in1k` | [`kerasformers/maxvit_large_tf_512_in21k_ft_in1k`](https://huggingface.co/kerasformers/maxvit_large_tf_512_in21k_ft_in1k) |
| `maxvit_small_tf_224_in1k` | [`kerasformers/maxvit_small_tf_224_in1k`](https://huggingface.co/kerasformers/maxvit_small_tf_224_in1k) |
| `maxvit_small_tf_384_in1k` | [`kerasformers/maxvit_small_tf_384_in1k`](https://huggingface.co/kerasformers/maxvit_small_tf_384_in1k) |
| `maxvit_small_tf_512_in1k` | [`kerasformers/maxvit_small_tf_512_in1k`](https://huggingface.co/kerasformers/maxvit_small_tf_512_in1k) |
| `maxvit_tiny_tf_224_in1k` | [`kerasformers/maxvit_tiny_tf_224_in1k`](https://huggingface.co/kerasformers/maxvit_tiny_tf_224_in1k) |
| `maxvit_tiny_tf_384_in1k` | [`kerasformers/maxvit_tiny_tf_384_in1k`](https://huggingface.co/kerasformers/maxvit_tiny_tf_384_in1k) |
| `maxvit_tiny_tf_512_in1k` | [`kerasformers/maxvit_tiny_tf_512_in1k`](https://huggingface.co/kerasformers/maxvit_tiny_tf_512_in1k) |
| `maxvit_xlarge_tf_224_in21k` | [`kerasformers/maxvit_xlarge_tf_224_in21k`](https://huggingface.co/kerasformers/maxvit_xlarge_tf_224_in21k) |
| `maxvit_xlarge_tf_384_in21k_ft_in1k` | [`kerasformers/maxvit_xlarge_tf_384_in21k_ft_in1k`](https://huggingface.co/kerasformers/maxvit_xlarge_tf_384_in21k_ft_in1k) |
| `maxvit_xlarge_tf_512_in21k_ft_in1k` | [`kerasformers/maxvit_xlarge_tf_512_in21k_ft_in1k`](https://huggingface.co/kerasformers/maxvit_xlarge_tf_512_in21k_ft_in1k) |

## Tips

- Set `KERAS_BACKEND` **before** importing Keras / kerasformers.
- `MaxViTImageClassify` returns class logits; `MaxViTModel` returns features (`as_backbone=True` for multi-scale stages).
- See [docs](https://imvision12.github.io/KerasFormers/classification_backbones/) and [Loading Weights](https://imvision12.github.io/KerasFormers/loading_weights/).
- Upstream / timm checkpoints: `MaxViTImageClassify.from_weights("hf:timm/maxvit_tiny_tf_512.in1k")`.

## Special Thanks

A huge thank you to the MaxViT authors and the timm / Hub communities for creating and releasing these models.

License: see YAML `license` (usually matches the upstream checkpoint).