11 repos
Tools and models for fine-tuning and serving Mistral language models, built around the safetensors format and transformer-compatible endpoints. The cluster includes specialized datasets (news and Reddit content from various regions) paired with infrastructure for model adaptation and inference, reflecting an applied focus on deploying customized Mistral instances across different domains and languages.