Update README.md
Browse files
README.md
CHANGED
|
@@ -5,9 +5,9 @@ language:
|
|
| 5 |
pipeline_tag: any-to-any
|
| 6 |
---
|
| 7 |
|
| 8 |
-
## **Spectra-561B-27B-
|
| 9 |
|
| 10 |
-
**Omnira Spectra-561B-27B-
|
| 11 |
---
|
| 12 |
|
| 13 |
### **Core Architecture**
|
|
@@ -48,7 +48,7 @@ The model utilizes a massive **Mixture-of-Experts (MoE)** structure that balance
|
|
| 48 |
| --- | --- |
|
| 49 |
| **Total Parameters** | 561B |
|
| 50 |
| **Activated Parameters** | 27B |
|
| 51 |
-
| **Expert Configuration** | 512 Routed Experts |
|
| 52 |
| **Context Window** | 128K tokens |
|
| 53 |
| **Primary Tasks** | Audio, Visual, Text, Video-Continuation |
|
| 54 |
|
|
@@ -73,20 +73,20 @@ Due to its massive scale (561B), Spectra requires multi-node clusters or high-me
|
|
| 73 |
```python
|
| 74 |
from spectra_omni import SpectraModel
|
| 75 |
|
| 76 |
-
model = SpectraModel.from_pretrained("thenexthub/Spectra-561B-27B-
|
| 77 |
# Seamlessly process audio, image, and text inputs
|
| 78 |
|
| 79 |
```
|
| 80 |
|
| 81 |
## License Agreement
|
| 82 |
|
| 83 |
-
The **model weights** are released under the **
|
| 84 |
|
| 85 |
## Citation
|
| 86 |
|
| 87 |
```
|
| 88 |
@misc{omnira2026spectra,
|
| 89 |
-
title={Spectra-561B-27B-
|
| 90 |
author={Omnira},
|
| 91 |
year={2026},
|
| 92 |
url={https://github.com/theomnira/Spectra-Omni},
|
|
|
|
| 5 |
pipeline_tag: any-to-any
|
| 6 |
---
|
| 7 |
|
| 8 |
+
## **Spectra-561B-27B-768E: Model Overview**
|
| 9 |
|
| 10 |
+
**Omnira Spectra-561B-27B-768E** is a state-of-the-art open-source omni-modal model developed by Omnira's Artificial Intelligence Team. Designed for **real-time audio-visual interaction**, it integrates text, image, video, and audio processing into a single, unified framework.
|
| 11 |
---
|
| 12 |
|
| 13 |
### **Core Architecture**
|
|
|
|
| 48 |
| --- | --- |
|
| 49 |
| **Total Parameters** | 561B |
|
| 50 |
| **Activated Parameters** | 27B |
|
| 51 |
+
| **Expert Configuration** | 512 Routed; 256 Zero Experts |
|
| 52 |
| **Context Window** | 128K tokens |
|
| 53 |
| **Primary Tasks** | Audio, Visual, Text, Video-Continuation |
|
| 54 |
|
|
|
|
| 73 |
```python
|
| 74 |
from spectra_omni import SpectraModel
|
| 75 |
|
| 76 |
+
model = SpectraModel.from_pretrained("thenexthub/Spectra-561B-27B-768E")
|
| 77 |
# Seamlessly process audio, image, and text inputs
|
| 78 |
|
| 79 |
```
|
| 80 |
|
| 81 |
## License Agreement
|
| 82 |
|
| 83 |
+
The **model weights** are released under the **Omnira License**. This license does not grant any rights to use Omnira trademarks or patents.
|
| 84 |
|
| 85 |
## Citation
|
| 86 |
|
| 87 |
```
|
| 88 |
@misc{omnira2026spectra,
|
| 89 |
+
title={Spectra-561B-27B-768E: Unified Omni-modal Intelligence},
|
| 90 |
author={Omnira},
|
| 91 |
year={2026},
|
| 92 |
url={https://github.com/theomnira/Spectra-Omni},
|