ms-marco-TinyBERT-L2-v2-F16-GGUF
This model was converted to GGUF format from cross-encoder/ms-marco-TinyBERT-L-2-v2 using llama.cpp via the ggml.ai's GGUF-my-repo space.
Refer to the original model card for more details on the model.
Model Information
- Base Model: cross-encoder/ms-marco-TinyBERT-L-2-v2
- Quantization: F16
- Format: GGUF (GPT-Generated Unified Format)
- Converted with: llama.cpp
Quantization Details
This is a F16 quantization of the original model:
- F16: Full 16-bit floating point - highest quality, largest size
- Q8_0: 8-bit quantization - high quality, good balance
- Q4_K_M: 4-bit quantization with medium quality - smaller size, faster inference
Usage
This model can be used with llama.cpp and other GGUF-compatible inference engines.
# Example using llama.cpp
./llama-rerank -m ms-marco-TinyBERT-L2-v2-F16.gguf
Model Files
| Quantization | Use Case |
|---|---|
| F16 | Maximum quality, largest size |
| Q8_0 | High quality, good balance of size/performance |
| Q4_K_M | Good quality, smallest size, fastest inference |
Citation
If you use this model, please cite the original model:
# See original model card for citation information
License
This model inherits the license from the original model. Please refer to the original model card for license details.
Acknowledgements
- Original model by the authors of cross-encoder/ms-marco-TinyBERT-L-2-v2
- GGUF conversion via llama.cpp by ggml.ai
- Converted and uploaded by sinjab