File size: 4,750 Bytes
cc5734b b51e234 cc5734b b51e234 cc5734b b51e234 cc5734b b51e234 cc5734b b51e234 cc5734b b51e234 cc5734b b51e234 cc5734b b51e234 cc5734b b51e234 cc5734b b51e234 cc5734b b51e234 cc5734b b51e234 cc5734b b51e234 cc5734b b51e234 cc5734b | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 | ---
pipeline_tag: any-to-any
license: apache-2.0
base_model: google/gemma-4-E2B
library_name: zeromodels
language:
- en
tags:
- keras
- zeromodels
- gemma4
- gemma-4
- any-to-any
- pytorch
- jax
- tf
---
*See [our collection](https://huggingface.co/zeromodels) for all Gemma 4 sizes and variants.*
# Run Gemma 4 with Keras 3: JAX, PyTorch, or TensorFlow
[](https://github.com/IMvision12/ZeroModels) [](https://imvision12.github.io/ZeroModels/gemma4/) [](https://huggingface.co/zeromodels)
# zeromodels/gemma-4-e2b
Pure-**Keras 3** conversion of [`google/gemma-4-E2B`](https://huggingface.co/google/gemma-4-E2B) for
[zeromodels](https://github.com/IMvision12/ZeroModels). One implementation runs unmodified on
**TensorFlow / Torch / JAX**. This is the **5B** variant, served here as
**image + audio + text -> text** via `Gemma4Processor`; weights are stored in **bfloat16**.
For model details, license, and usage terms, see Google's
[model card](https://huggingface.co/google/gemma-4-E2B).
## Gemma 4 family
| Property | E2B | E4B | 12B Unified | 31B Dense |
| --- | --- | --- | --- | --- |
| Total Parameters | 2.3B effective (5.1B with embeddings) | 4.5B effective (8B with embeddings) | 11.95B | 30.7B |
| Layers | 35 | 42 | 48 | 60 |
| Sliding Window | 512 tokens | 512 tokens | 1024 tokens | 1024 tokens |
| Context Length | 128K tokens | 128K tokens | 256K tokens | 256K tokens |
| Vocabulary Size | 262K | 262K | 262K | 262K |
| Supported Modalities | Text, Image, Audio | Text, Image, Audio | Text, Image, Audio | Text, Image |
| Vision Encoder Parameters | ~150M | ~150M | - | ~550M |
| Audio Encoder Parameters | ~300M | ~300M | - | No Audio |
## ✨ Quick start
### Text-only
```python
import os
os.environ["KERAS_BACKEND"] = "torch" # or "jax" / "tensorflow"
from zeromodels.models.gemma4 import Gemma4TextGenerate, Gemma4Tokenizer
model = Gemma4TextGenerate.from_weights("zeromodels/gemma-4-e2b")
tokenizer = Gemma4Tokenizer.from_weights("zeromodels/gemma-4-e2b")
inputs = tokenizer([{"role": "user", "content": "Hello, who are you?"}])
outputs = model.generate(**inputs, max_new_tokens=64)
print(tokenizer.decode(outputs[0]))
```
### Image + audio + text
```python
import os
os.environ["KERAS_BACKEND"] = "torch" # or "jax" / "tensorflow"
from PIL import Image
from zeromodels.models.gemma4 import Gemma4ConditionalGenerate, Gemma4Processor
model = Gemma4ConditionalGenerate.from_weights("zeromodels/gemma-4-e2b")
processor = Gemma4Processor.from_weights("zeromodels/gemma-4-e2b")
inputs = processor(conversation=[
{"role": "user", "content": [
{"type": "image", "image": Image.open("cat.jpg")},
{"type": "audio", "path": "clip.wav"},
{"type": "text", "text": "Describe the image and what you hear."},
]}
])
outputs = model.generate(**inputs, max_new_tokens=64)
print(processor.decode(outputs[0]))
```
Load any Gemma 4 variant the same way with `from_weights("zeromodels/<variant>")`:
| Variant | Hub |
| --- | --- |
| `gemma-4-12b` | [zeromodels/gemma-4-12b](https://huggingface.co/zeromodels/gemma-4-12b) |
| `gemma-4-12b-it` | [zeromodels/gemma-4-12b-it](https://huggingface.co/zeromodels/gemma-4-12b-it) |
| `gemma-4-26b-a4b` | [zeromodels/gemma-4-26b-a4b](https://huggingface.co/zeromodels/gemma-4-26b-a4b) |
| `gemma-4-26b-a4b-it` | [zeromodels/gemma-4-26b-a4b-it](https://huggingface.co/zeromodels/gemma-4-26b-a4b-it) |
| `gemma-4-31b` | [zeromodels/gemma-4-31b](https://huggingface.co/zeromodels/gemma-4-31b) |
| `gemma-4-31b-it` | [zeromodels/gemma-4-31b-it](https://huggingface.co/zeromodels/gemma-4-31b-it) |
| `gemma-4-e2b` | [zeromodels/gemma-4-e2b](https://huggingface.co/zeromodels/gemma-4-e2b) |
| `gemma-4-e2b-it` | [zeromodels/gemma-4-e2b-it](https://huggingface.co/zeromodels/gemma-4-e2b-it) |
| `gemma-4-e4b` | [zeromodels/gemma-4-e4b](https://huggingface.co/zeromodels/gemma-4-e4b) |
| `gemma-4-e4b-it` | [zeromodels/gemma-4-e4b-it](https://huggingface.co/zeromodels/gemma-4-e4b-it) |
## Tips
- Set `KERAS_BACKEND` **before** importing Keras / zeromodels.
- Loads in **bfloat16** by default. Pass `load_dtype="float32"` for full precision,
or `quantization="int8"` to shrink further.
- See the [Gemma 4 docs](https://imvision12.github.io/ZeroModels/gemma4/).
- Community / upstream weights still work via the `hf:` prefix:
`Gemma4ConditionalGenerate.from_weights("hf:google/gemma-4-E2B")`.
## Special Thanks
A huge thank you to the Google Gemma authors for creating and releasing these models.
License: Apache 2.0.
|