Axolotl
Named by 4 kept items in this workspace.
Search every workspace for Axolotl. It matches the name as written, so it can find other things that share it.
The items that name it 4
Fine-tuning tool with day-0 LoRA support for Mistral Large 4, using NVFP4 experts and merge-aware LoRA on one 8xB200 node.
Fine-tuning tool that now trains merge-aware so RL LoRAs merged into BF16 weights match what vLLM serves.
LoRA training framework built on transformers, could train LoRA directly over GGUF models
LoRA training framework built on transformers that could benefit from native GGUF support
All technologies in this workspace.