Articles from simonwillison.net

Articles from simonwillison.net

The term “context engineering” is gaining traction over “prompt engineering” as it better describes the skill of providing LLMs with the necessary information (simonwillison.net)
2 days ago

OpenAI updates its coding agent Codex with internet access, turned off by default, and expands availability to ChatGPT Plus users (simonwillison.net)
26 days ago

Large Language Models can run tools in your terminal with LLM 0.26 (simonwillison.net)
2025-5-28

Mistral launches an API for agents, which can run code, make images, access docs, search the web, and “hand off” to other agents, similar to OpenAI's offerings (simonwillison.net)
2025-5-28

Researchers detail an exploit in GitHub's official MCP server that lets hackers trick an LLM agent into leaking private information about the MCP user (simonwillison.net)
2025-5-27 GitHub

A database tracking instances where lawyers got caught presenting AI hallucinations shows that, of 116 cases dating back to June 2023, 20 occurred this month (simonwillison.net)
2025-5-27 Database AI

Highlights from the system prompts of Claude Opus 4 and Claude Sonnet 4, including model safety, avoiding sycophancy, and not regurgitating copyrighted content (simonwillison.net)
2025-5-26

In April ChatGPT started to, by default, reference all past chats for more personalized responses, but this means users lose control of their prompts' context (simonwillison.net)
2025-5-22

Understanding the recent criticism of the Chatbot Arena (simonwillison.net)
2025-5-1 chatbot

Giving software away for free (simonwillison.net)
2025-4-29

Watching OpenAI's o3 guess a photo's location feels surreal, dystopian, and entertaining, including running Python code to examine details like license plates (simonwillison.net)
2025-4-27

Image segmentation using Gemini 2.5 (simonwillison.net)
2025-4-18

CaMeL offers a promising new direction for mitigating prompt injection attacks (simonwillison.net)
2025-4-12

Long context support in LLM 0.24 using fragments and template plugins (simonwillison.net)
2025-4-8

Gemini 2.5 Pro hands-on: a very strong model with 1M input and 64K output tokens, a January 2025 knowledge cut-off, and very, very impressive coding skills (simonwillison.net)
2025-3-26

In an addendum to the GPT-4o system card, OpenAI says it is not blocking the image generation of adult public figures and that public figures can opt out (simonwillison.net)
2025-3-26

Alibaba releases Qwen2.5-VL-32B, a 32B open model under Apache 2.0, claims better alignment with human preferences and math reasoning than earlier 2.5 VL models (simonwillison.net)
2025-3-25 Alibaba Mathematics

DeepSeek releases MIT-licensed DeepSeek-V3-0324, the latest version of their enormous DeepSeek v3 model; the previous DeepSeek v3 version had a custom license (simonwillison.net)
2025-3-25

Here's how I use LLMs to help me write code (simonwillison.net)
2025-3-12

Will the future of software development run on vibes? (simonwillison.net)
2025-3-11

Structured data extraction from unstructured content using LLM schemas (simonwillison.net)
2025-3-1

A look at Claude 3.7 Sonnet's extended thinking mode and its 128K token output limit; long thinking runs are impressive but can take several minutes to complete (simonwillison.net)
2025-2-26

Run LLMs on macOS using llm-mlx and Apple's MLX framework (simonwillison.net)
2025-2-16 Apple

shot-scraper 1.6 with support for HTTP Archives (simonwillison.net)
2025-2-14 web crawler

A selfish personal argument for releasing code as Open Source (simonwillison.net)
2025-2-3 Open Source

OpenAI's o3-mini costs $1.10 per 1M input tokens and $4.40 per 1M output tokens, cheaper than GPT-4o, which costs $2.50 and $10, and o1, which costs $15 and $60 (simonwillison.net)
2025-2-1

OpenAI updates ChatGPT's Canvas feature with o1 model support and HTML and React code rendering, making it a direct competitor to Claude's Artifacts (simonwillison.net)
2025-1-25

Chinese AI lab DeepSeek debuts DeepSeek-R1, an MIT-licensed model that does well with math, code, and reasoning tasks, alongside other open and distilled models (simonwillison.net)
2025-1-21 China AI Mathematics

I still don't think companies serve you ads based on spying through your microphone (simonwillison.net)
2025-1-3

Things we learned out about LLMs in 2024 (simonwillison.net)
2025-1-1

Previous Page Next Page