Instructions to use VextLabsinc/gem-ruby with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use VextLabsinc/gem-ruby with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("VextLabsinc/juwel-beryl") model = PeftModel.from_pretrained(base_model, "VextLabsinc/gem-ruby") - Notebooks
- Google Colab
- Kaggle
NOTICE
Browse files
NOTICE
CHANGED
|
@@ -1,10 +1,17 @@
|
|
| 1 |
-
NOTICE —
|
| 2 |
|
| 3 |
Copyright 2026 Vext Labs, Inc.
|
| 4 |
|
| 5 |
-
This
|
| 6 |
-
|
| 7 |
-
|
|
|
|
| 8 |
|
| 9 |
-
|
| 10 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
NOTICE — JUWEL GEM open-weight adapter (code)
|
| 2 |
|
| 3 |
Copyright 2026 Vext Labs, Inc.
|
| 4 |
|
| 5 |
+
This is a LoRA adapter for the JUWEL base line (JUWEL Sapphire / JUWEL Emerald),
|
| 6 |
+
which is derived from Qwen3-VL-32B-Instruct by Alibaba Cloud (Tongyi Qianwen Lab),
|
| 7 |
+
released under the Apache License 2.0. Original base model:
|
| 8 |
+
https://huggingface.co/Qwen/Qwen3-VL-32B-Instruct
|
| 9 |
|
| 10 |
+
Modifications by Vext Labs, Inc.: refusal-direction abliteration, CIP organic
|
| 11 |
+
upscale (additional transformer layers, trained), and this domain LoRA adapter.
|
| 12 |
+
|
| 13 |
+
Attribution only — the "Qwen"/"Alibaba" names and marks are used solely to satisfy
|
| 14 |
+
Apache-2.0 attribution and do NOT imply endorsement by Alibaba Cloud.
|
| 15 |
+
|
| 16 |
+
Apache-2.0 requires retaining this NOTICE and the LICENSE file in redistributions,
|
| 17 |
+
and stating that changes were made (they were — see above).
|