Laboratory / GPUs for training, inference and fine-tuning
IteraGPU / ML research
What task are you giving the model?
Training, inference and adaptation require different trade-offs between memory, precision and duration. Start from your workload to prepare the right configuration and protocol.
GPU training: validate the loop before the campaign
Prepare the data, the batch and the resumption of a training run. One protocol per phase to control memory, gradients, validation and output files.
Prepare the protocolGPU inference: choosing based on memory, context and quality
Size inference with the context, the concurrency, the memory peak and the output quality. A concrete protocol before choosing your GPU package.
Prepare the protocolFine-tuning: choosing an adaptation and verifying its contribution
Prepare a fine-tuning with a reference, separate data, and a reload check. LoRA, quantized base, budget, and regression analysis.
Prepare the protocol