Models Can Prompt Now

Posted4 months agoActive4 months ago

KolenCh

6 points

3 comments

simonwillison.netTechstory

calmpositive

Debate

20/100

LlmsAIPrompt Engineering

Key topics

Llms

Prompt Engineering

The article discusses how models can now generate effective prompts, and the HN discussion explores the implications and examples of this development, including its potential applications and the role of human evaluation.

Snapshot generated from the HN discussion

Discussion Activity

Light discussion

First comment

Peak period

0-1h

Avg / period

1.5

Key moments

01Story posted
Sep 14, 2025 at 5:23 PM EDT
4 months ago
Step 01
02First comment
Sep 14, 2025 at 5:29 PM EDT
6m after posting
Step 02
03Peak activity
2 comments in 0-1h
Hottest window of the conversation
Step 03
04Latest activity
Sep 14, 2025 at 6:59 PM EDT
4 months ago
Step 04

Generating AI Summary...

Analyzing up to 500 comments to identify key contributors and discussion patterns

Discussion (3 comments)

Showing 3 comments

iguana2000

4 months ago

Makes sense. This is probably one of the few clear examples of models getting better by learning from content online written about themselves.

vunderba

4 months ago

This has been a thing for a while now - particularly around early versions of image generation models. It wasn't entirely uncommon to wire up a small llama-based LLM model into a ComfyUI workflow that would enrich and/or rewrite simpler prompts using SD 1.5 style notation ("hyperrealistic, octane render, 8k, etc").

KolenChAuthor

4 months ago

My wife is doing a freelance project where basically humans are evaluating how good an LLM-generated prompt is, and if there’s any problem to be fixed. Some times the humans are amazed and can’t believe they are written by LLMs (other times they are flabbergasted by how bad it is, of course.)

So definitely it has been improved in this area, intentionally.

View full discussion on Hacker News

ID: 45243359Type: storyLast synced: 11/17/2025, 2:04:05 PM

Want the full context?

Jump to the original sources

Read the primary article or dive into the live Hacker News thread when you're ready.

Open link View on HN