| Property | Value |
|---|---|
| Parameter Count | 32.5B |
| License | Other |
| Format | GGUF |
| Author | tensorblock |
What is WizardLM-30B-Uncensored-Guanaco-SuperCOT-30b-GGUF?
This is a powerful uncensored language model converted to GGUF format, offering various quantization options for different performance and resource requirements. The model is based on WizardLM-30B and has been trained on seven diverse datasets including WizardLM alpaca, SuperCOT, and Guanaco datasets.
Implementation Details
The model is available in multiple quantization levels ranging from Q2_K (11.2GB) to Q8_0 (32.2GB), allowing users to choose between file size and quality tradeoffs. The recommended versions are Q4_K_M for balanced performance and Q5_K_M for higher quality with reasonable size requirements.
- Compatible with llama.cpp (commit b4011)
- Multiple quantization options for different use cases
- Optimized for resource efficiency while maintaining quality
Core Capabilities
- Uncensored text generation and completion
- Enhanced performance through SuperCOT dataset integration
- Flexible deployment options with various GGUF quantization levels
- Balanced quality-to-size ratio options
Frequently Asked Questions
Q: What makes this model unique?
This model combines the power of WizardLM with uncensored capabilities and SuperCOT enhancement, while offering multiple quantization options for various deployment scenarios. The GGUF format makes it highly compatible with modern inference engines.
Q: What are the recommended use cases?
The model is suitable for applications requiring unrestricted language generation, with different quantization options serving various use cases - from resource-constrained environments (Q2_K) to high-quality requirements (Q6_K, Q8_0).