Simon Willison’s Weblog

Newsletters: 2025

1. 

   28 December 2025 · Substack

   A new way to extract detailed transcripts from Claude Code

2. 

   17 December 2025 · Substack

   OpenAI are quietly adopting skills, now available in ChatGPT and Codex CLI

3. 

   12 December 2025 · Substack

   GPT 5.2 and useful patterns for building HTML tools

4. 1 December 2025 · Monthly newsletter

   LLM digest: November 2025

5. 

   25 November 2025 · Substack

   Claude Opus 4.5, and why evaluating new LLMs is increasingly difficult

6. 

   19 November 2025 · Substack

   Trying out Gemini 3 Pro with audio transcription and a new pelican benchmark

7. 

   11 November 2025 · Substack

   Code research projects with async coding agents like Claude Code and Codex

8. 

   6 November 2025 · Substack

   A new SQL-powered permissions system in Datasette 1.0a20

9. 

   3 November 2025 · Substack

   New prompt injection papers: Agents Rule of Two and The Attacker Moves Second

10. 1 November 2025 · Monthly newsletter

    LLM digest: October 2025

11. 

    24 October 2025 · Substack

    Building a tool to copy-paste share terminal sessions using Claude Code for web

12. 

    21 October 2025 · Substack

    Claude Code for web—a new asynchronous coding agent from Anthropic

13. 

    17 October 2025 · Substack

    Claude Skills are awesome, maybe a bigger deal than MCP

14. 

    15 October 2025 · Substack

    NVIDIA DGX Spark: great hardware, early days for the ecosystem

15. 

    8 October 2025 · Substack

    Vibe engineering

16. 

    6 October 2025 · Substack

    Embracing the parallel coding agent lifestyle

17. 

    2 October 2025 · Substack

    Designing agentic loops

18. 1 October 2025 · Monthly newsletter

    LLM digest: September 2025

19. 

    30 September 2025 · Substack

    Claude Sonnet 4.5 is probably the “best coding model in the world” (at least for now)

20. 

    25 September 2025 · Substack

    GPT-5-Codex, plus updated Gemini 2.5 Flash and Flash Lite

21. 

    18 September 2025 · Substack

    I think “agent” may finally have a widely enough agreed upon definition to be useful jargon now

22. 

    9 September 2025 · Substack

    Claude’s new Code Interpreter

23. 

    7 September 2025 · Substack

    GPT-5 Thinking in ChatGPT (aka Research Goblin) is shockingly good at search

24. 

    2 September 2025 · Substack

    V&A East Storehouse and Operation Mincemeat in London

25. 1 September 2025 · Monthly newsletter

    LLM digest: August 2025

26. 

    21 August 2025 · Substack

    Prompt injections as far as the eye can see

27. 

    12 August 2025 · Substack

    LLM 0.27, with GPT-5 and improved tool calling

28. 

    8 August 2025 · Substack

    GPT-5: Key characteristics, pricing and model card

29. 

    1 August 2025 · Substack

    Reverse engineering some updates to Claude

30. 1 August 2025 · Monthly newsletter

    LLM digest: July 2025

31. 

    30 July 2025 · Substack

    My 2.5 year old laptop can write Space Invaders in JavaScript now, using GLM-4.5 Air and MLX

32. 

    26 July 2025 · Substack

    Using GitHub Spark to reverse engineer GitHub Spark

33. 

    18 July 2025 · Substack

    Vibe scraping on an iPhone with OpenAI Codex

34. 

    11 July 2025 · Substack

    Grok 4: searching X for “from:elonmusk (Israel OR Palestine OR Hamas OR Gaza)”

35. 

    1 July 2025 · Substack

    Gemma 3n, Context Engineering and a whole lot of Claude Code

36. 1 July 2025 · Monthly newsletter

    LLM digest: June 2025

37. 

    26 June 2025 · Substack

    Phoenix.new is Fly’s entry into the prompt-driven app development space

38. 

    21 June 2025 · Substack

    Trying out the new Gemini 2.5 model family

39. 

    17 June 2025 · Substack

    The lethal trifecta for AI agents

40. 

    11 June 2025 · Substack

    Comma v0.1 1T and 2T—7B LLMs trained on openly licensed text

41. 

    6 June 2025 · Substack

    The last six months in LLMs, illustrated by pelicans on bicycles

42. 

    31 May 2025 · Substack

    How often do LLMs snitch? Recreating Theo’s SnitchBench with LLM

43. 30 May 2025 · Monthly newsletter

    LLM digest: May 2025

44. 

    29 May 2025 · Substack

    Large Language Models can run tools in your terminal with LLM 0.26

45. 

    26 May 2025 · Substack

    Highlights from the Claude 4 system prompt

46. 

    25 May 2025 · Substack

    Building software on top of Large Language Models

47. 

    11 May 2025 · Substack

    Trying out llama.cpp’s new vision support

48. 

    6 May 2025 · Substack

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

49. 

    29 April 2025 · Substack

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

50. 

    22 April 2025 · Substack

    AI assisted search-based research actually works now

51. 

    18 April 2025 · Substack

    Image segmentation using Gemini 2.5

52. 

    14 April 2025 · Substack

    CaMeL offers a promising new direction for mitigating prompt injection attacks

53. 

    11 April 2025 · Substack

    Model Context Protocol has prompt injection security problems

54. 

    8 April 2025 · Substack

    Long context support in LLM 0.24 using fragments

55. 

    4 April 2025 · Substack

    Gemini 2.5 Pro pricing, a new GPT-4o

56. 

    26 March 2025 · Substack

    Putting Gemini 2.5 Pro through its paces

57. 

    25 March 2025 · Substack

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

58. 

    19 March 2025 · Substack

    Not all AI-assisted programming is vibe coding (but vibe coding rocks)

59. 

    14 March 2025 · Substack

    Notes on Google’s Gemma 3

60. 

    11 March 2025 · Substack

    How I use LLMs to help me write code

61. 

    9 March 2025 · Substack

    What’s new in the world of LLMs, for NICAR 2025

62. 

    5 March 2025 · Substack

    Hallucinations in code are the least dangerous form of LLM mistakes

63. 

    28 February 2025 · Substack

    Structured data extraction from unstructured content using LLM schemas

64. 

    25 February 2025 · Substack

    Claude 3.7 Sonnet, extended thinking and long output

65. 

    17 February 2025 · Substack

    Run LLMs on macOS using llm-mlx and Apple’s MLX framework

66. 

    14 February 2025 · Substack

    URL-addressable Pyodide Python environments

67. 

    2 February 2025 · Substack

    OpenAI o3-mini, now available in LLM

68. 

    29 January 2025 · Substack

    A selfish personal argument for releasing code as Open Source

69. 

    24 January 2025 · Substack

    Anthropic’s new Citations API

70. 

    20 January 2025 · Substack

    The DeepSeek-R1 family of reasoning models