65 repos across 2 sub-areas
Mistral-based Text Generation Models
35 repos
Fine-tuned and quantized variants of the Mistral 7B language model, optimized for different tasks including general text generation, code generation, and instruction-following. The cluster consists primarily of model repositories and their GGUF quantized versions, representing a collection of specialized adaptations of the base Mistral architecture with different training objectives (code-focused, multi-task, conversational) and inference optimizations for resource-constrained deployment.
Large Language Models for Code Generation
30 repos
Text generation and code synthesis using transformer-based large language models, primarily centered on CodeLlama variants and related inference infrastructure. These repositories focus on pre-trained and instruction-tuned models optimized for code understanding and generation tasks, along with the inference frameworks and endpoints needed to deploy them. Researchers and practitioners exploring this cluster will find model implementations, fine-tuning approaches, and production deployment patterns for code-specific LLMs.