57 lines
1.9 KiB
Markdown
57 lines
1.9 KiB
Markdown
---
|
|
base_model: CraneAILabs/swahili-gemma-1b
|
|
language:
|
|
- en
|
|
library_name: transformers
|
|
pipeline_tag: text-generation
|
|
quantized_by: ThijsL202
|
|
tags:
|
|
- gguf
|
|
- quantized
|
|
- llama-cpp
|
|
|
|
---
|
|
|
|
# CraneAILabs_swahili-gemma-1b - GGUF Standard
|
|
|
|
## 📊 Quantization Details
|
|
|
|
- **Base Model**: [CraneAILabs/swahili-gemma-1b](https://huggingface.co/CraneAILabs/swahili-gemma-1b)
|
|
- **Quantization**: Standard
|
|
- **Total Size**: 4.03 GB (5 files)
|
|
|
|
### 📦 Standard Quantizations Classic GGUF quantizations without importance matrix enhancement.
|
|
|
|
## 📁 Available Files
|
|
|
|
| Quantization | Size | Download |
|
|
|:------------|-----:|:---------:|
|
|
| **Q2_K** | 690MB | [⬇️](https://huggingface.co/ThijsL202/CraneAILabs_swahili-gemma-1b-GGUF/resolve/main/CraneAILabs_swahili-gemma-1b.Q2_K.gguf) |
|
|
| **Q3_K_L** | 752MB | [⬇️](https://huggingface.co/ThijsL202/CraneAILabs_swahili-gemma-1b-GGUF/resolve/main/CraneAILabs_swahili-gemma-1b.Q3_K_L.gguf) |
|
|
| **Q4_K_M** | 806MB | [⬇️](https://huggingface.co/ThijsL202/CraneAILabs_swahili-gemma-1b-GGUF/resolve/main/CraneAILabs_swahili-gemma-1b.Q4_K_M.gguf) |
|
|
| **Q6_K** | 1.0GB | [⬇️](https://huggingface.co/ThijsL202/CraneAILabs_swahili-gemma-1b-GGUF/resolve/main/CraneAILabs_swahili-gemma-1b.Q6_K.gguf) |
|
|
| **Q8_0** | 1.1GB | [⬇️](https://huggingface.co/ThijsL202/CraneAILabs_swahili-gemma-1b-GGUF/resolve/main/CraneAILabs_swahili-gemma-1b.Q8_0.gguf) |
|
|
|
|
## 🚀 Quick Start
|
|
|
|
### llama.cpp
|
|
```bash
|
|
./llama-cli -m CraneAILabs_swahili-gemma-1b.Q6_K.gguf -p "Your prompt here" -n 512
|
|
```
|
|
|
|
### Python
|
|
```python
|
|
from llama_cpp import Llama
|
|
|
|
llm = Llama(model_path="./CraneAILabs_swahili-gemma-1b.Q6_K.gguf", n_ctx=2048)
|
|
output = llm("Your prompt here", max_tokens=512)
|
|
print(output["choices"][0]["text"])
|
|
```
|
|
|
|
## 📊 Model Information
|
|
|
|
Original model: [CraneAILabs/swahili-gemma-1b](https://huggingface.co/CraneAILabs/swahili-gemma-1b)
|
|
|
|
---
|
|
*Quantized using [llama.cpp](https://github.com/ggml-org/llama.cpp)*
|