Cluster 647079

Large language models from Meta's Llama family, including various quantized and optimized versions across different model sizes (1B through 70B parameters). This cluster contains model checkpoints and implementations for running Llama models efficiently, spanning formats like GGML and GPTQ quantization, enabling deployment on resource-constrained hardware. Developers exploring this cluster will find pre-trained model weights, inference optimization techniques, and tooling for serving and fine-tuning open-source large language models.

Meta's Llama 2 family of open-source large language models, spanning multiple parameter scales (7B, 13B, 70B) with both base and chat-tuned variants. The cluster centers on model checkpoints, weights, and implementations optimized for inference and fine-tuning using PyTorch. This collection serves researchers and practitioners exploring production-ready open LLMs and their deployment across different computational requirements.

Fine-tuned and adapted transformer-based language models optimized for German text generation and NLP tasks. This cluster centers on variants of base models (7B, 13B, and 70B parameters) tailored for German language understanding, including both standard PyTorch implementations and quantized GGUF formats for efficient inference. Repositories here represent different versions and optimizations of German-specialized models built on transformer architectures.