Skip to content

armgit / Soup

Public Imported from GitHub
0

Fine-tune LLMs from one YAML. Layer streaming trains an 8B model on a 4 GB laptop GPU.

509 B
Raw Blame
# Soup template: Vision / Multimodal
# Fine-tune a vision-language model for image understanding

base: meta-llama/Llama-3.2-11B-Vision-Instruct
task: sft
modality: vision
backend: transformers  # The vision SFT path does not use Unsloth.

data:
  train: ./data/vision_train.jsonl
  format: llava
  image_dir: ./data/images
  val_split: 0.1
  max_length: 2048

training:
  epochs: 3
  lr: 1e-5
  batch_size: auto
  lora:
    r: 64
    alpha: 16
    target_modules: auto
  quantization: 4bit

output: ./output