Mistral 7B Model Variants & GGUF Quantization

11 repos

Quantized and fine-tuned versions of the Mistral 7B language model in GGUF format, designed for efficient inference on consumer hardware. These repositories represent various experimental model merges and specialized adaptations (including German language variants and instruction-tuned versions) that leverage model merging techniques like DARE to combine capabilities from multiple base models. Developers exploring this cluster will find pretrained model weights, quantization approaches, and text generation configurations for running capable open-weight LLMs with reduced computational requirements.