Post-training, explained with numbers.
Field notes from the team that runs thousands of fine-tuning jobs a month. No hype, no hand-waving: methods, memory maths, costs and the eval practices that keep releases honest.
SFT vs DPO vs GRPO: choosing a post-training method
When supervised fine-tuning is enough, when preference optimization pays off, and when reinforcement learning with verifiable rewards is worth the compute.
LoRA vs full fine-tuning: quality, cost and serving trade-offs
Rank, target modules, memory math and where adapters stop being enough — with real numbers from 8B to 70B.
How much does it cost to fine-tune an LLM?
A line-item breakdown of GPU hours, data, evals and serving, with worked examples for 8B, 32B and 70B models.
Evaluating fine-tuned models: how to test your evals
Golden sets, LLM-as-judge calibration, contamination checks and the regression gates that keep a release honest.
How to fine-tune an LLM on Databricks
Unity Catalog data, governed GPU clusters, MLflow lineage and Model Serving — the end-to-end pattern for post-training inside your lakehouse.
Choosing a post-training platform: a buyer's guide
Hosted fine-tuning APIs, managed platforms and DIY stacks compared on control, cost, security and time-to-production.
Bring us a model and a metric.
Tell us what you want the model to do better. A post-training engineer will come back within one business day with a plan, a compute estimate and an eval design.