# HN: Show HN: Minimal LLM Post-Training Experiments on an 8GB GPU (SFT, DPO, GRPO) — retained record

> Retained hackernews record from hackernews.

Canonical: https://highsignal.app/data/records/eea95ae1cb69ff50

[← sources](https://highsignal.app/data)


source: [hackernews](https://highsignal.app/data/hackernews) published 2026-08-01 12:30 ingested 2026-08-02 08:17

[original source ↗](https://news.ycombinator.com/item?id=49133851)

## retained text



```
Points: 21 | Comments: 0
Author: popopanda
Link: https://github.com/pochenai/nano-llm-posttraining
Show HN: Minimal LLM Post-Training Experiments on an 8GB GPU (SFT, DPO, GRPO)
```



document fetched 2026-08-08 06:41
