OZZZER · AI NEWS1 of 3 free stories opened
← Back to AI News

AI Hardware & Infrastructure · 19 Nov 2026 · 01:39 CET

AWS Brings Full-Weight Training To Serverless SageMaker Customization

Unite.AI · 19 Nov 2026 · 01:39 CETRead original at Unite.AI ↗
Share
LinkedInXFacebookWhatsApp
AWS Brings Full-Weight Training To Serverless SageMaker Customization

Publisher preview · OZZZER analysis pending editorial review.

PUBLISHER ARTICLE PREVIEW

From the original article

Amazon Web Services announced on November 18, 2026 full fine-tuning (FFT) support for SageMaker AI Serverless Model Customization, and reported 90.0% semantic correctness in a coding experiment using more than 100,000 training examples, while cautioning that aggressive settings caused severe general-knowledge losses.

Customers supply a base model, dataset, and hyperparameters; the service allocates compute and manages checkpoints without requiring GPU provisioning, cluster setup, or distributed-training configuration. In AWS’s SageMaker Python SDK example, switching methods means replacing TrainingType.FULL with TrainingType.LORA, leaving the rest of the job configuration unchanged.

FFT updates every model weight. LoRA instead freezes pretrained weights and inserts low-rank adapter matrices, reducing GPU capacity requirements but limiting updates to a smaller representational space. AWS’s explanation is that full-rank training can capture more complex domain adaptations as datasets grow, rather than simply increasing the number of trainable parameters.

AWS compared the methods using supervised fine-tuning (SFT) of a Qwen3-4B base model, with identical prompts, targets, and evaluation code. The Magicoder-Evol-Instruct-110K dataset supplied coding instruction-response pairs covering algorithm design, debugging, data structures, and systems programming. The split assigned 99,000 records to training, 5,500 to validation, and 5,500 to a held-out test set.

The evaluations separated task performance from knowledge retention. Amazon Nova Pro served as an LLM judge, reading generated code with the original prompt to assess logical correctness.

Source

Unite.AI · 19 Nov 2026 · 01:39 CET

Open the original at Unite.AI ↗