Update README.md
Browse files
README.md
CHANGED
@@ -19,6 +19,8 @@ If you have questions, please feel free to reach out to us on
|
|
19 |
Learning Daily [Discord
|
20 |
community](https://discord.com/invite/p9ecgRhDR8/).
|
21 |
|
|
|
|
|
22 |
## **[DeciLM 6B](https://huggingface.co/Deci/DeciLM-6b):**
|
23 |
A 5.7 billion parameter, permissively licensed, pretrained LLM using variable Grouped-Query Attention (GQA) to achieve an optimal balance between performance and computational efficiency. Generated using Deci's proprietary Neural Architecture Search technology, AutoNAC, DeciLM 6B delivers a throughput that’s 15 times faster than Llama 2 7B while maintaining comparable quality.
|
24 |
DeciLM 6B-Instruct, fine-tuned from DeciLM 6B for instruction-following use cases, is among the most accurate open-source models in the 7 billion parameter class.
|
@@ -37,6 +39,10 @@ Neural Architecture Search technology. YOLO-NAS is a game-changer in the
|
|
37 |
world of object detection, providing superior real-time object detection
|
38 |
capabilities and production-ready performance.
|
39 |
|
|
|
|
|
|
|
|
|
40 |
## **[SuperGradients](https://github.com/Deci-AI/super-gradients):**
|
41 |
An open-source library for training PyTorch-based computer vision
|
42 |
models. Developed by Deci's deep learning experts for the benefit of the
|
@@ -54,7 +60,7 @@ user-friendly report.
|
|
54 |
Simplify and accelerate the development of computer vision and NLP and
|
55 |
Generative AI applications with highly accurate and efficient foundation
|
56 |
models and advanced tools to customize, optimize, and deploy them to
|
57 |
-
production.
|
58 |
|
59 |
Deci is powered by groundbreaking Automated Neural Architecture
|
60 |
Construction (AutoNAC™) technology. Deci's AutoNAC™ engine democratizes
|
|
|
19 |
Learning Daily [Discord
|
20 |
community](https://discord.com/invite/p9ecgRhDR8/).
|
21 |
|
22 |
+
In addition to releasing top-notch models, Deci recently launched Infery-LLM, the most advanced inference SDK for LLM optimization and deployment,
|
23 |
+
|
24 |
## **[DeciLM 6B](https://huggingface.co/Deci/DeciLM-6b):**
|
25 |
A 5.7 billion parameter, permissively licensed, pretrained LLM using variable Grouped-Query Attention (GQA) to achieve an optimal balance between performance and computational efficiency. Generated using Deci's proprietary Neural Architecture Search technology, AutoNAC, DeciLM 6B delivers a throughput that’s 15 times faster than Llama 2 7B while maintaining comparable quality.
|
26 |
DeciLM 6B-Instruct, fine-tuned from DeciLM 6B for instruction-following use cases, is among the most accurate open-source models in the 7 billion parameter class.
|
|
|
39 |
world of object detection, providing superior real-time object detection
|
40 |
capabilities and production-ready performance.
|
41 |
|
42 |
+
## **[Infery-LLM](https://deci.ai/infery-llm-book-a-demo/):**
|
43 |
+
The most advanced inference SDK for LLM optimization and deployment, Infery-LLM includes unique features such as optimized kernels, continuous batching, advanced selective quantization, ultra-efficient beam search, parallel execution, and more.
|
44 |
+
Book a [personalized demo](https://deci.ai/infery-llm-book-a-demo/) to explore Infery-LLM's capabilities.
|
45 |
+
|
46 |
## **[SuperGradients](https://github.com/Deci-AI/super-gradients):**
|
47 |
An open-source library for training PyTorch-based computer vision
|
48 |
models. Developed by Deci's deep learning experts for the benefit of the
|
|
|
60 |
Simplify and accelerate the development of computer vision and NLP and
|
61 |
Generative AI applications with highly accurate and efficient foundation
|
62 |
models and advanced tools to customize, optimize, and deploy them to
|
63 |
+
production.
|
64 |
|
65 |
Deci is powered by groundbreaking Automated Neural Architecture
|
66 |
Construction (AutoNAC™) technology. Deci's AutoNAC™ engine democratizes
|