---
base_model: Sao10K/L3-8B-Stheno-v3.2
tags:
- text-generation-inference
- transformers
- unsloth
- llama
- trl
- sft
license: apache-2.0
language:
- en
- zh
---

# TigerStheno

- **Developed by:** safe049
- **License:** apache-2.0
- **Finetuned from model :** Sao10K/L3-8B-Stheno-v3.2

This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.

[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)

### Model Details
**Model Name:** TigerStheno-8B-V3.2  
**Permalink:** [TigerStheno-8B-V3.2](https://huggingface.co/TigerResearch/TigerStheno-8B-V3.2)  
**Model Description:**  
TigerStheno-8B-V3.2 is an AI model based on the Sao10K/L3-8B-Stheno-v3.2 architecture, fine-tuned using the TigerBot dataset. This model is designed to enhance language understanding and generation capabilities, particularly in multi-language and multi-task role-playing scenarios.  

- **Developed by:** safe049  
- **Shared by:** safe049  
- **Model type:** Large Language Model  
- **Language(s) (NLP):** Multi-language support  
- **License:** Apache 2.0  
- **Finetuned from model:** Sao10K/L3-8B-Stheno-v3.2  

### Uses
**Direct Use:**  
The model can be directly used for various natural language processing tasks such as role-playing, text generation, summarization, translation, and dialogue systems.  
 

### Bias, Risks
- **Bias:** TigerStheno-8B-V3.2 is uncensored and may give any response including illegal,in-moral content. 
- **Risks:** May generate illegal and in-moral content

### Using it
**GGUF:**
Download the gguf file in the repo, and use it in any of these apps and etc:
- KoboldCPP
- Ollama
- LlamaCPP
etc.

**Transformer**
Here is a example code snippet to use it with transformer:
```python
from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "TigerResearch/TigerStheno-8B-V3.2"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)

# Example input
inputs = tokenizer("Hello, how are you?", return_tensors="pt")
outputs = model.generate(**inputs)
print(tokenizer.decode(outputs[0]))
```