Simon Willison's Weblog: Entries simonwillison.net

I really don't like ChatGPT's new memory feature

simonwillison.net·51w·Hacker News

Building software on top of Large Language Models

simonwillison.net·51w·Lobsters, Hacker News, Hacker News

Trying out llama.cpp's new vision support

simonwillison.net·52w·Hacker News

Saying "hi" to Microsoft's Phi-4-reasoning

simonwillison.net·53w

Feed a video to a vision LLM as a sequence of JPEG frames on the CLI (also LLM 0.25)

simonwillison.net·53w

Two publishers and three authors fail to understand what "vibe coding" means

simonwillison.net·53w·Hacker News

Understanding the recent criticism of the Chatbot Arena

simonwillison.net·54w·Lobsters, Hacker News

Qwen 3 offers a case study in how to effectively release a model

simonwillison.net·54w·Hacker News

Watching o3 guess a photo's location is surreal, dystopian and wildly entertaining

simonwillison.net·54w·Lobsters, Hacker News

Exploring Promptfoo via Dave Guarino's SNAP evals

simonwillison.net·54w·Hacker News

AI assisted search-based research actually works now

simonwillison.net·55w·Hacker News

Maybe Meta's Llama claims to be open source because of the EU AI act

simonwillison.net·55w·Hacker News, Hacker News

Image segmentation using Gemini 2.5

simonwillison.net·55w·Lobsters, Hacker News

GPT-4.1: Three new million token input models from OpenAI, including their cheapest model yet

simonwillison.net·56w

CaMeL offers a promising new direction for mitigating prompt injection attacks

simonwillison.net·56w·Lobsters, Hacker News

Model Context Protocol has prompt injection security problems

simonwillison.net·57w·Hacker News

Long context support in LLM 0.24 using fragments and template plugins

simonwillison.net·57w·Lobsters, Hacker News

Note on 5th April 2025

simonwillison.net·57w·Hacker News

Putting Gemini 2.5 Pro through its paces

simonwillison.net·59w·Hacker News

New audio models from OpenAI, but how much can we rely on them?

simonwillison.net·59w·Hacker News

Log in to enable infinite scrolling