Instructions to use sassoftware/gemma-4-31B-it-assistant with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use sassoftware/gemma-4-31B-it-assistant with Transformers:
# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("sassoftware/gemma-4-31B-it-assistant") model = AutoModelForCausalLM.from_pretrained("sassoftware/gemma-4-31B-it-assistant", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Add SAS Note to Model Card
#1
by nadolsw - opened
README.md
CHANGED
|
@@ -562,4 +562,21 @@ The development of vision-language models (VLMs) raises several ethical concerns
|
|
| 562 |
|
| 563 |
### **Benefits**
|
| 564 |
|
| 565 |
-
At the time of release, this family of models provides high-performance open vision-language model implementations designed from the ground up for responsible AI development compared to similarly sized models.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 562 |
|
| 563 |
### **Benefits**
|
| 564 |
|
| 565 |
+
At the time of release, this family of models provides high-performance open vision-language model implementations designed from the ground up for responsible AI development compared to similarly sized models.
|
| 566 |
+
|
| 567 |
+
## Source Model Repo
|
| 568 |
+
|
| 569 |
+
This model is derived from [`google/gemma-4-31B-it-assistant`](https://huggingface.co/google/gemma-4-31B-it-assistant).
|
| 570 |
+
See the upstream repository for the original safetensors weights, training data, and the full upstream model card.
|
| 571 |
+
|
| 572 |
+
## SAS Modifications
|
| 573 |
+
|
| 574 |
+
SAS has not made any modifications to the source model weights or file artifacts.
|
| 575 |
+
|
| 576 |
+
## SAS Product Integration
|
| 577 |
+
|
| 578 |
+
This repo is consumed by the SAS Document Analysis product offering
|
| 579 |
+
|
| 580 |
+
## Model Security Scan
|
| 581 |
+
|
| 582 |
+
Model weights were scanned on 2026-06-22. All checks passed. Safetensor file received a LOW risk level assessment.
|