Update README.md
Browse files
README.md
CHANGED
@@ -7,4 +7,39 @@ sdk: static
|
|
7 |
pinned: false
|
8 |
---
|
9 |
|
10 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
7 |
pinned: false
|
8 |
---
|
9 |
|
10 |
+
# Persian-llm-fibonacci-1-7b-chat.P1_0 π
|
11 |
+
|
12 |
+
## Description π
|
13 |
+
The **Persian-llm-fibonacci-1-7b-chat.P1_0** is a **1.7 billion parameter language model (LLM)** specifically designed for **Persian-language chat and text interactions**. Developed as part of the **FibonacciAI** project, this model is optimized to generate fluent and natural Persian text, making it ideal for conversational AI applications.
|
14 |
+
|
15 |
+
Built on advanced language model architectures (e.g., GPT), it excels in tasks like chat, content generation, question answering, and more. π
|
16 |
+
|
17 |
+
---
|
18 |
+
|
19 |
+
## Use Cases π‘
|
20 |
+
- **Chatbots**: Create intelligent Persian-language chatbots. π€
|
21 |
+
- **Content Generation**: Generate creative and contextually relevant Persian text. π
|
22 |
+
- **Question Answering**: Provide natural and accurate answers to user queries. β
|
23 |
+
- **Machine Translation**: Translate text to and from Persian. π
|
24 |
+
|
25 |
+
---
|
26 |
+
|
27 |
+
## How to Use π οΈ
|
28 |
+
To use this model, you can leverage the `transformers` library. Here's a quick example:
|
29 |
+
|
30 |
+
```python
|
31 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer
|
32 |
+
|
33 |
+
# Load the model and tokenizer
|
34 |
+
model_name = "fibonacciai/Persian-llm-fibonacci-1-7b-chat.P1_0"
|
35 |
+
model = AutoModelForCausalLM.from_pretrained(model_name)
|
36 |
+
tokenizer = AutoTokenizer.from_pretrained(model_name)
|
37 |
+
|
38 |
+
# Generate a response to an input text
|
39 |
+
input_text = "Ψ³ΩΨ§Ω
Ψ ΪΨ·ΩΨ±ΫΨ"
|
40 |
+
inputs = tokenizer(input_text, return_tensors="pt")
|
41 |
+
outputs = model.generate(**inputs, max_length=50)
|
42 |
+
|
43 |
+
# Decode the output to text
|
44 |
+
response = tokenizer.decode(outputs[0], skip_special_tokens=True)
|
45 |
+
print(response)
|