Update README.md
Browse files
README.md
CHANGED
@@ -5,18 +5,31 @@ pipeline_tag: text-generation
|
|
5 |
extra_gated_button_content: Acknowledge license
|
6 |
tags:
|
7 |
- conversational
|
|
|
|
|
|
|
8 |
---
|
9 |
|
10 |
|
11 |
-
#
|
12 |
|
13 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
14 |
|
15 |
-
## π Our Flagship Model: Silma 1.0 π
|
16 |
-
**Silma 1.0** is the **TOP-RANKED** Arabic LLM with an impressive **9 billion parameter size**, surpassing models that are over seven times larger. π
|
17 |
|
18 |
## π₯ Our Team
|
19 |
-
|
|
|
20 |
|
21 |
**Authors**: [silma.ai](https://silma.ai)
|
22 |
|
@@ -385,6 +398,4 @@ Risks identified and mitigations:
|
|
385 |
and application use cases.
|
386 |
* Privacy violations: Models were trained on data filtered for removal of PII
|
387 |
(Personally Identifiable Information). Developers are encouraged to adhere to
|
388 |
-
privacy regulations with privacy-preserving techniques.
|
389 |
-
|
390 |
-
|
|
|
5 |
extra_gated_button_content: Acknowledge license
|
6 |
tags:
|
7 |
- conversational
|
8 |
+
language:
|
9 |
+
- ar
|
10 |
+
- en
|
11 |
---
|
12 |
|
13 |
|
14 |
+
# SILMA AI
|
15 |
|
16 |
+
SILMA.AI is a leading Generative AI startup dedicated to empowering Arabic speakers with state-of-the-art AI solutions.
|
17 |
+
|
18 |
+
|
19 |
+
## π Our Flagship Model: SILMA 1.0 π
|
20 |
+
|
21 |
+
* **SILMA 1.0** is the **TOP-RANKED** open-weights Arabic LLM with an impressive **9 billion parameter size**, surpassing models that are over seven times larger. π
|
22 |
+
|
23 |
+
## What makes SILMA exceptional?
|
24 |
+
|
25 |
+
* SIMLA is a small language model outperforming 72B models in most arabic language tasks, thus more practical for business use-cases
|
26 |
+
* SILMA is built on the strong foundational models of Google Gemma, giving you the best of both worlds
|
27 |
+
* SILMA is an open-weight model free to use in accordance with our open license
|
28 |
|
|
|
|
|
29 |
|
30 |
## π₯ Our Team
|
31 |
+
|
32 |
+
We are a team of seasoned **Arabic AI experts** who understand the nuances of the language and cultural considerations, enabling us to build solutions that truly resonate with Arabic users.
|
33 |
|
34 |
**Authors**: [silma.ai](https://silma.ai)
|
35 |
|
|
|
398 |
and application use cases.
|
399 |
* Privacy violations: Models were trained on data filtered for removal of PII
|
400 |
(Personally Identifiable Information). Developers are encouraged to adhere to
|
401 |
+
privacy regulations with privacy-preserving techniques.
|
|
|
|