doberst commited on
Commit
dde9797
1 Parent(s): c16333e

Upload README.md

Browse files
Files changed (1) hide show
  1. README.md +16 -24
README.md CHANGED
@@ -1,37 +1,29 @@
1
  ---
2
- license: llama2
3
- inference: false
4
- tags: [green, llmware-rag, p7, ov]
5
  ---
6
 
7
- # dragon-llama-2-ov
8
 
9
- <!-- Provide a quick summary of what the model is/does. -->
10
-
11
- **dragon-llama-2-ov** is an OpenVino int4 quantized version of Dragon Llama-2, providing a very fast, very small inference implementation, optimized for AI PCs using Intel GPU, CPU and NPU.
12
-
13
- [**dragon-llama-2**](https://huggingface.co/llmware/dragon-llama-2-7b-v0) is a fact-based question-answering model, optimized for complex business documents.
14
-
15
- Get started right away with [OpenVino](https://github.com/openvinotoolkit/openvino)
16
-
17
- Looking for AI PC solutions and demos, contact us at [llmware](https://www.llmware.ai)
18
 
 
19
 
20
  ### Model Description
21
 
22
  - **Developed by:** llmware
23
- - **Model type:** llama2
24
- - **Parameters:** 7 billion
25
- - **Model Parent:** llmware/dragon-llama-2-v0
 
26
  - **Language(s) (NLP):** English
27
- - **License:** llama2-community-license
28
- - **Uses:** Fact-based question-answering
29
- - **RAG Benchmark Accuracy Score:** 97.25
30
- - **Quantization:** int4
31
-
32
 
33
- ## Model Card Contact
34
 
 
 
35
  [llmware on hf](https://www.huggingface.co/llmware)
36
-
37
- [llmware website](https://www.llmware.ai)
 
1
  ---
2
+ license: apache-2.0
3
+ inference: false
4
+ tags: [green, llmware-rag, p1, ov]
5
  ---
6
 
7
+ # bling-tiny-llama-ov
8
 
9
+ **bling-tiny-llama-ov** is a very small, very fast fact-based question-answering model, designed for retrieval augmented generation (RAG) with complex business documents, quantized and packaged in OpenVino int4 for AI PCs using Intel GPU, CPU and NPU.
 
 
 
 
 
 
 
 
10
 
11
+ This model is one of the smallest and fastest in the series. For higher accuracy, look at larger models in the BLING/DRAGON series.
12
 
13
  ### Model Description
14
 
15
  - **Developed by:** llmware
16
+ - **Model type:** tinyllama
17
+ - **Parameters:** 1.1 billion
18
+ - **Quantization:** int4
19
+ - **Model Parent:** [llmware/bling-tiny-llama-v0](https://www.huggingface.co/llmware/bling-tiny-llama-v0)
20
  - **Language(s) (NLP):** English
21
+ - **License:** Apache 2.0
22
+ - **Uses:** Fact-based question-answering, RAG
23
+ - **RAG Benchmark Accuracy Score:** 86.5
 
 
24
 
 
25
 
26
+ ## Model Card Contact
27
+ [llmware on github](https://www.github.com/llmware-ai/llmware)
28
  [llmware on hf](https://www.huggingface.co/llmware)
29
+ [llmware website](https://www.llmware.ai)