Citrus1.0-Qwen-72B-i1-GGUF
| Property | Value |
|---|---|
| Base Model | Citrus1.0-Qwen-72B |
| Model Type | GGUF Quantized |
| Author | mradermacher |
| Size Range | 22.8GB - 64.4GB |
| Repository | Hugging Face |
What is Citrus1.0-Qwen-72B-i1-GGUF?
This is a specialized quantized version of the Citrus1.0-Qwen-72B model, offering various GGUF formats optimized for different deployment scenarios. The model utilizes weighted/imatrix quantization techniques to provide multiple compression options while maintaining performance.
Implementation Details
The model offers a comprehensive range of quantization options, from highly compressed IQ1_S (22.8GB) to high-quality Q6_K (64.4GB) variants. The implementation features both standard and imatrix-based quantization methods, with IQ-quants often providing better performance than similarly-sized standard quants.
- Multiple quantization levels (IQ1, IQ2, IQ3, IQ4, Q4, Q5, Q6)
- Various size options for each quantization level
- Optimized balance between file size and model quality
- GGUF format for efficient deployment
Core Capabilities
- Flexible deployment options with different size-quality tradeoffs
- Recommended Q4_K_M variant for optimal performance (47.5GB)
- Support for both high-compression (IQ1_S) and high-quality (Q6_K) use cases
- Efficient memory usage through various quantization techniques
Frequently Asked Questions
Q: What makes this model unique?
This model stands out for its comprehensive range of quantization options and the use of innovative imatrix quantization techniques, offering better performance than traditional quantization methods at similar sizes.
Q: What are the recommended use cases?
For optimal balance between performance and size, the Q4_K_M variant (47.5GB) is recommended. For systems with limited resources, IQ3 variants offer good performance at smaller sizes, while Q6_K provides near-original model quality for high-end applications.
