1
Nano Banana can be prompt engineered for nuanced AI image generation
Nano Banana allows 32,768 input tokens and I’m going to try to use them all dammit.
Max Woolf
A blog written by data scientist Max Woolf out of San Francisco.
- Popular posts
- 40
- Total points
- 10,047
- HN comments
- 4,640
- Rank
- #82 personal
Visit minimaxir.com → RSS feedOn Hacker News Oct 2012 – Oct 2026
AI & Machine Learning 19Databases & Data 15Software Engineering 11Tech Industry & Business 8Games & Graphics 6Society, Policy & Culture 5
2
Can LLMs write better code if you keep asking them to “write better code”?
Most coders want AI to write code faster: I want AI to write FASTER CODE.
3
ChatGPT's API is so good and cheap, it makes most text generating AI obsolete
Including OpenAI’s other text generating AI!
4
OpenAI and Anthropic cut their LLM prices by up to 80% within a few months. Why?
What is “the true cost of a token”?
5
As an experienced LLM user, I don't use generative LLMs often
But for what I do use LLMs for, it’s invaluable.
6
I Made Stable Diffusion XL Smarter by Finetuning It on Bad AI-Generated Images
And then telling it to not generate those images!
7
Pushing ChatGPT's Structured Data Support to Its Limits
“Function calling” with ChatGPT is ChatGPT’s best feature since ChatGPT.
8
I Resurrected “Ugly Sonic” with Stable Diffusion Textual Inversion
The true dystopia of AI is that it can make Ugly Sonic look cool.
9
The Problem with LangChain
LangChain is complicated, so it must be better. Right?
10
Tempering Expectations for GPT-3 and OpenAI’s API
GPT-3 is indeed a large step forward for AI text-generation, but there are very many caveats with the popular demos and use cases.
11
Does offering ChatGPT a tip cause it to generate better text?
Modern AI rewards being very weird.
12
Benchmarking TensorFlow on Cloud CPUs: Cheaper Deep Learning Than Cloud GPUs
Using CPUs instead of GPUs for deep learning training in the cloud is cheaper because of the massive cost differential afforded by preemptible instances.
13
How to create a blog post title optimizer with GPT-3 and Hacker News data
GPT-3 says the title for this post is very bad.
14
Stable Diffusion 2.0 and the Importance of Negative Prompts for Good Results
Negative prompts can be far superior than traditional prompt additions.
15
Fun and dystopia with AI-based code generation using GPT-J-6B
At the least, AI-generated code is much more readable than the average human’s.
16
The super effectiveness of Pokémon embeddings using only raw JSON and images
Embeddings encourage engineers to go full YOLO because it’s actually rewarding to do so!
17
18
19
Writing Rust code that's fast by asking agents to make the code faster
c’mon, try doing a breakthrough