|
|
--- |
|
|
license: mit |
|
|
language: |
|
|
- en |
|
|
- fr |
|
|
- de |
|
|
- es |
|
|
- it |
|
|
- pt |
|
|
- ja |
|
|
- ko |
|
|
- zh |
|
|
- ar |
|
|
- el |
|
|
- fa |
|
|
- pl |
|
|
- id |
|
|
- cs |
|
|
- he |
|
|
- hi |
|
|
- nl |
|
|
- ro |
|
|
- ru |
|
|
- tr |
|
|
- uk |
|
|
- vi |
|
|
datasets: |
|
|
- xmanii/mauxi-talk-pro |
|
|
- xmanii/mauxitalk-persian |
|
|
--- |
|
|
|
|
|
# Hormoz 8B |
|
|
|
|
|
## Introduction |
|
|
|
|
|
This model is an effort in order to make a multi-lingual and _on device_ models which can be executed on the consumer hardware. The model follows the steps used in training _DeepSeek_ model. However, the model is _not a reasoning model_ and a generic question answering, conversational and _uncensored_ model which has been made with a cost of around $4000 USD. |
|
|
|
|
|
If you're curious about the model you also can see our [GitHub](https://github.com/mann-e/hormoz) and learn more about the benchmarks and costs. |
|
|
|
|
|
Also, this model is based on _Command R_'s architecture, since that architecture gave us the best results in multilingual chat. Specially with languages such as _Persian_ and _Arabic_. This way, you can consider this model like a commercially useaeble version of _aya expanse_ as well. |
|
|
|
|
|
### The name |
|
|
|
|
|
<p align="center"> |
|
|
<img src="https://github.com/Mann-E/hormoz/blob/main/hormoz-logo.png?raw=true" width=768px /> |
|
|
</p> |
|
|
|
|
|
The name __Hormoz__ comes from the Persian word "هرمز" which has multiple meanings. It can point to the _strait of Hormoz_ in Persian Gulf or _Hormoz Island_ which is part of the Hormozgan Province in the south of Iran. Also it may point to "اورمزد" or _Ourmozd_ which is middle/ancient Persian name for the planet _Jupiter_ and derived from the term _Ahura Mazda_ or the Avestan term for God. |
|
|
|
|
|
## How to run (transformers) |
|
|
|
|
|
### Free API |
|
|
|
|
|
The model is also available through [Jabir Project's API](https://jabirproject.org/api-docs) and [Pollinations.AI](https://pollinations.ai). |
|
|
|
|
|
### Install transformers |
|
|
|
|
|
``` |
|
|
pip install transformers --upgrade |
|
|
``` |
|
|
|
|
|
_Note:_ For better performance, you may need to install `accelerate` package as well. |
|
|
|
|
|
### Inference |
|
|
|
|
|
```python |
|
|
from transformers import AutoTokenizer, AutoModelForCausalLM |
|
|
|
|
|
model_id = "mann-e/Hormoz-8B" |
|
|
tokenizer = AutoTokenizer.from_pretrained(model_id) |
|
|
model = AutoModelForCausalLM.from_pretrained(model_id).to("cuda") |
|
|
|
|
|
messages = [{"role": "user", "content": "What is the answer to universe, life and everything?"}] |
|
|
input_ids = tokenizer.apply_chat_template(messages, tokenize=True, add_generation_prompt=True, return_tensors="pt").to("cuda") |
|
|
|
|
|
gen_tokens = model.generate( |
|
|
input_ids, |
|
|
max_new_tokens=1024, |
|
|
do_sample=True, |
|
|
temperature=1.0, |
|
|
) |
|
|
|
|
|
gen_text = tokenizer.decode(gen_tokens[0]) |
|
|
print(gen_text) |
|
|
``` |
|
|
|
|
|
## License |
|
|
|
|
|
This model is published under _MIT_ license. |
|
|
|
|
|
### Commercial Use |
|
|
|
|
|
Since this model is MIT licensed, you're free to do whatever you want with the model. However since we're a relatively small startup, we recommend you if you are a big corporate and you host this model, give us a capacity of your API as well. This way, we both can benefit from the model. |
