simonw

Follow

Simon Willison simonw

Follow

4.9k followers · 136 following

Datasette
Half Moon Bay, CA
18:43 (UTC -07:00)
https://simonwillison.net/
@simonw
@simon@fedi.simonwillison.net

Sponsors

Sponsoring

Achievements

BetaSend feedback

Achievements

BetaSend feedback

Highlights

Developer Program Member
Pro

Block or Report

Block or report simonw

Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse

simonw/README.md

Currently working on Datasette and associated projects. Read my blog or follow @simon@simonwillison.net on Mastodon.

Recent releases

datasette-export-notebook 1.0.1 - 2023-09-15

llm-sentence-transformers 0.1.2 - 2023-09-13

llm-clip 0.1 - 2023-09-12

llm 0.10 - 2023-09-12

datasette-cluster-map 0.18a0 - 2023-09-11

sqlite-chronicle 0.1 - 2023-09-11

blip-caption 0.1 - 2023-09-10

sqlite-utils 3.35.1 - 2023-09-09

More recent releases

On my blog

Weeknotes: Embeddings, more embeddings and Datasette Cloud - 2023-09-17

Build an image search engine with llm-clip, chat with models with llm chat - 2023-09-12

LLM now provides tools for working with embeddings - 2023-09-04

Datasette 1.0a4 and 1.0a5, plus weeknotes - 2023-08-30

Making Large Language Models work for you - 2023-08-27

Datasette Cloud, Datasette 1.0a3, llm-mlc and more - 2023-08-16

More on simonwillison.net

TIL

Limited JSON API for Google searches using Programmable Search Engine - 2023-09-17

Running tests against multiple versions of a Python dependency in GitHub Actions - 2023-09-15

Using llama-cpp-python grammars to generate JSON - 2023-09-13

Trying out cr-sqlite on macOS - 2023-09-13

Summarizing Hacker News discussion themes with Claude and LLM - 2023-09-09

Embedding paragraphs from my blog with E5-large-v2 - 2023-09-08

More on til.simonwillison.net

Pinned

datasette Public

An open source multi-tool for exploring and publishing data

Python 8.2k 588
sqlite-utils Public

Python CLI utility and library for manipulating SQLite databases

Python 1.3k 89
llm Public

Access large language models from the command-line

Python 1.2k 49
symbex Public

Find the Python code for specified symbols

Python 155 4
s3-credentials Public

A tool for creating credentials for accessing S3 buckets

Python 158 11
shot-scraper Public

A command-line utility for taking automated screenshots of websites

Python 1.1k 54

5,236 contributions in the last year

Contribution Graph
	Sep		Oct					Nov				Dec				Jan					Feb				Mar				Apr					May				Jun				Jul					Aug				Sep
Sun
Mon
Tue
Wed
Thu
Fri
Sat

Learn how we count contributions

Activity overview

Contributed to simonw/llm, simonw/datasette, simonw/til and 289 other repositories

Contribution activity

September 2023

Created 259 commits in 30 repositories

Created 11 repositories

simonw/usearch C++ Sep 13
simonw/llm-clip Python Sep 12
simonw/sqlite-chronicle Python Sep 11
simonw/blip-caption Python Sep 10
simonw/datasette-llm-embed Python Sep 8
simonw/bert-for-laptops Jupyter Notebook Sep 8
simonw/code-interpreter-data Sep 6
simonw/cr-sqlite Rust Sep 5
simonw/llm-cluster Python Sep 4
simonw/llm-sentence-transformers Python Sep 2
simonw/sqlite-utils-move-tables Python Sep 1

Created a pull request in simonw/llm that received 5 comments

Binary embeddings

Refs: #253 Still needed: Decide what to do about --store - should it store the binary content in a new content_blob column? See these notes. The

+228 −59 • 5 comments

Opened 15 other pull requests in 9 repositories

simonw/datasette 3 merged

actors_from_ids plugin hook and datasette.actors_from_ids() method Sep 8
Test against Python 3.12 preview Sep 6
click-default-group>=1.2.3 Sep 6

simonw/llm 3 merged

unum-cloud/usearch 2 merged

Fix typo: accerlation -> acceleration Sep 13
Remove error-causing annotation in README Sep 13

simonw/sqlite-utils 1 merged 1 open

simonw/datasette-export-notebook 1 merged

Upgrade for compatibility with Datasette 1.0 alphas Sep 15

datasette/datasette-write-ui 1 open

Potential fix for clipped menu bug Sep 9

simonw/ttok 1 open

Support for other models via Hugging Face tokenizers Sep 9

simonw/datasette-edit-schema 1 merged

Edit foreign and primary keys Sep 7

vlcn-io/cr-sqlite 1 merged

Typo fix, receied -> received Sep 5

Reviewed 2 pull requests in 2 repositories

simonw/datasette 1 pull request

datasette.yaml plugin support Sep 13

datasette/datasette-write-ui 1 pull request

Potential fix for clipped menu bug Sep 9

Created an issue in simonw/sqlite-chronicle that received 16 comments

Try out synchronization approaches

There are two synchronization cases to consider: A fresh sync where you need to copy the entire table An update, where you just want the rows that…

16 comments

Opened 106 other issues in 32 repositories

simonw/llm 34 closed 12 open

README should show off local models Sep 13
llm.user_dir() should create it if it doesn't exist yet Sep 13
Idea: pipe data to llm chat to start the conversation Sep 13
Fix test failures from non-stable ordering Sep 12
Ship 0.10, then ship a llm-clip non-alpha Sep 12
llm similar --binary option - for finding images similar to an image Sep 12
str(response) should output response.text() Sep 12
Multi-line input for llm chat Sep 12
llm collections should default to llm collections list Sep 12
Binary embeddings final cleanup Sep 12
llm collections show collection_id command Sep 10
Add llm plugins --all option to view default plugins, as seen in Datasette Sep 10
Make tests robust against extra plugins Sep 10
Environment variable to disable plugins Sep 10
Support for embedding binary files Sep 9
llm chat -c fails for API key models Sep 6
Research ways to speed up brute force cosine similarity Sep 6
llm chat -o options Sep 5
llm chat command Sep 5
Duplicate the --save feature from openai-to-sqlite similar Sep 5
Promote collections to a top-level command Sep 5
Document (or make easier) how to have Python code log to SQLite Sep 5
How could LLM handle fill-in-the-middle? Sep 4
Handle rate limiting for ada-002 embedding model Sep 4
llm embed-multi --files should handle encodings other than utf-8 Sep 4
Some issues not shown.

simonw/llm-sentence-transformers 8 closed

Error if io.datasette.llm folder does not exist yet Sep 13
AttributeError: 'SentenceTransformerModel' object has no attribute 'encode Sep 4
Got an error passing a generator to collection.embed_multi() Sep 3
Rename to llm-sentence-transformers Sep 2
Get --alias working Sep 2
Initial documentation Sep 2
Test a model in CI Sep 2
Initial plugin Sep 2

simonw/datasette 3 open 3 closed

simonw/airtable-export 3 closed 2 open

Save the Airtable schema to the output folder Sep 17
Option to export and save images and attachments Sep 17
Make stuff easier to view in Datasette Sep 17
Handle tables with / in the name Sep 17
Ability to export all tables in a base Sep 17

simonw/sqlite-utils 3 open 1 closed

simonw/datasette-edit-schema 4 closed

simonw/sqlite-chronicle 1 open 2 closed

simonw/datasette-graphql 3 open

simonw/llm-cluster 3 closed

Don't depend on sentence-transformers for tests Sep 4
Make truncation configurable Sep 4
Initial plugin design Sep 4

simonw/symbex 2 closed

simonw/datasette-packages 1 open

In-memory database of packages instead Sep 17

simonw/datasette-plugin 1 open

Port to pyproject.toml Sep 16

simonw/latest-datasette-with-all-plugins 1 closed

500 errors with Datasette 1.0 alpha Sep 15

mlc-ai/mlc-llm 1 open

[Feature Request] Python packages on PyPI Sep 13

simonw/llm-clip 1 closed

Initial plugin Sep 12

pallets/click 1 open

progressbar(hide=True) option for hiding the progress bar Sep 12

simonw/datasette-cluster-map 1 open

Compatibility with Datasette 1.0 alphas Sep 11

datasette/datasette-sqlite-trace 1 closed

Initial plugin Sep 10

datasette/datasette-debug-permissions 1 open

Try using a wrapper=True hook for this Sep 10

simonw/blip-caption 1 closed

Initial design Sep 10

simonw/ttok 1 open

Ability to count tokens for models other than OpenAI Sep 9

dogsheep/dogsheep-beta 1 open

Run rebuild_fts after building the index Sep 8

simonw/datasette.io 1 closed

Search is broken Sep 8

simonw/datasette-llm-embed 1 open

Initial design Sep 8

datasette/datasette-debug-actors-from-ids 1 closed

Initial plugin design Sep 8

7 repositories not shown

69 contributions in private repositories Sep 1 – Sep 16