Upload README.md with huggingface_hub

Upload tokenizer_config.json with huggingface_hub

Update model card with benchmark results (MMLU 48.7%, HellaSwag 58%, GSM8K 55%)

Upload tokenizer_config.json with huggingface_hub

Upload tokenization_moonshot.py with huggingface_hub

Upload config.json with huggingface_hub

Upload configuration_deepseek.py with huggingface_hub

Upload modeling_deepseek.py with huggingface_hub

Upload config.json with huggingface_hub

Add requirements.txt for inference endpoint

Add custom inference handler for DeepSeekV3 architecture

Add model card

Upload Moonlight-16B-A3B-Instruct abliterated with Bruno MoE gate abliteration

initial commit