soft-shell crabvietnam crab exporter

Simon Willison’s Weblog

Newsletter

Tuesday, 6th October 2026

I wanted to see if Claude Opus 5.5 could compose music, so I tried this:

I want you to write some computer game music for me. First design simple text based format for the music and build an artifact that can play it out loud - include some example tracks in that artifact

I am looking for music of the quality of the original secret of Monkey Island

It leaned a lot harder into the Monkey Island theme than I had intended, but the results are surprisingly good.

Screenshot of a retro pixel-art web music player. Header in blackletter type reads "Scrimshaw Jukebox", with the text "Six original adventure-game tracks, written as plain text and played by a synthesizer running in your browser. Pick a tune, press Play, then open the score and change it." A large pixel-art scene shows a harbor at night under a purple starry sky: a full moon at top right reflecting on the water, an island silhouette on the left with palm trees and a hut with two lit windows, and a sailing ship moored at a wooden pier. Overlaid on the scene: "Moonlit Harbor" and "Press Play". Below it a bar reads "Play Moonlit Harbor". A control panel has buttons "Play", "Stop", "Restart", "Loop: on", "Edit score", "Read guide" and a "Volume" slider set to about three quarters. A track list of six cards, the first highlighted: "Moonlit Harbor 100 bpm · 4/4 · 16 voices · 1:26", "The Rusty Anchor 112 bpm · 6/8 · 8 voices · 0:56", "The Ghost Galleon 66 bpm · 4/4 · 9 voices · 2:11", "The Jungle Path 92 bpm · 4/4 · 12 voices · 1:29", "Duel on the Docks 152 bpm · 4/4 · 12 voices · 1:13", "Lantern Waltz 96 bpm · 3/4 · 8 voices · 1:38". A section titled "Score view" with the caption "1:26 · 4/4 at 100 bpm · Main theme. A calypso for a harbour town after dark." shows a piano-roll visualization of colored horizontal note bars and percussion ticks on a dark background, with a section marker "A" and a yellow vertical playhead line. A color-coded legend of voices reads: "pan steeldrum", "flute flute", "marimba marimba", "skank organ", "strings strings", "harp harp", "bass fretless", "timp timpani", "kick kick", "rim rim", "shaker shaker", "conga conga", "tumba tumba", "bongo bongo", "crash crash", "surf surf". Footer text: "Click a voice to mute it. Space bar plays and stops."

I wonder if the ability to compose competent music is similar to the 3D graphics thing - a new capability for text models that emerged in the past few months?

Would need some careful experiments with other recent and not-so-recent models to confirm if this is new or if they've been able to do this for a while.

A minor fix for compatibility with the latest Datasette alphas. This meant we could upgrade the datasette.io site to Datasette 1.0a41.

Comment My comment on Mistral Large 4 — Hacker News

wren6991: The benchmark is saturated. Frontier models are tested with an armadillo in fishnet tights jaywalking on Mars.

OK well I couldn't resist this one:

llm -m claude-opus-5.5 'Generate an SVG of an armadillo in fishnet tights jaywalking on Mars'
llm -m gpt-6.1-sol 'Generate an SVG of an armadillo in fishnet tights jaywalking on Mars'
llm -m gemini-3.8-flash 'Generate an SVG of an armadillo in fishnet tights jaywalking on Mars'
llm -m mistral/mistral-large-4 'Generate an SVG of an armadillo in fishnet tights jaywalking on Mars'

Default reasoning levels for each: https://tools.simonwillison.net/markdown-svg-renderer?url=ht...

# 6:20 pm

None

I saw Parseable in a Show HN today - it's a new observability platform with both an open source (AGPL) Rust implementation (a single ~180MB binary), an "Enterprise" version with extra features and a cloud hosted option.

Since Datasette 1.0a41 added OpenTelemetry support (thanks, Alex Garcia), I decided to fire up Codex and have it figure out how to run Parseable and feed it traces from Datasette.

Here's my (human-written) TIL showing the patterns that worked, and here's a screenshot of a Datasette trace displayed within the Parseable localhost web application:

Screenshot of a trace detail view in an observability web app, with a span waterfall overlaid on a dimmed navigation sidebar and filter column. Dimmed sidebar: breadcrumb "Community > Traces > datas" (cut off), search box "Search... ⌘K", nav items "Home", "Ingest telemetry", section "ANALYZE": "Keystone", "Dashboards", "SQL Editor", section "OBSERVE": "Logs", "Metrics", "Traces" (selected), "APM", "Agents", section "MONITOR": "Alerts", "Errors", section "DATA": "Datasets", and at the bottom "Settings", "Book a call", "Support". Dimmed filter column, cut off at the right edge: "Search fi", "Core", "Log format", "User agent", "Source IPs", "Error", "Service", "service.ins", "service.na", "datasett" (checked), "Span", "Database", "HTTP", "http.reque", "NULL" (unchecked), "GET" (checked), "http.respo", "http.route", "Server", "server.add", "Telemetry", "URL", "All fields". Trace panel header: "Trace detail > 6f819a2170bcd1e91c6ea3ae236ec60b" with a copy icon, a "Related logs" button and a close X. Summary: "Start time 6:56 PM, Oct 6, 2026 UTC", "Duration 40.9 ms", "Spans 247". A minimap with axis "0ns 10.2ms 20.5ms 30.7ms 40.9ms" shows many short span bars cascading diagonally from top left toward the lower middle, with a few longer bars. Below is a span table with a "Span name" header, a "Search spans..." box, collapse and expand buttons, and a timeline axis "0ns 10.2ms 20.5ms 30.7ms 40.9ms". Rows (name, service, duration): root span with collapse toggle "123", "GET /..." "datasette..." 40.9ms spanning the full timeline; then alternating rows where each "db.query" has a collapse toggle "1": db.query datasette-local 679µs, db.query.execute datasette-local 278µs, db.query datasette-local 341µs, db.query.execute datasette-local 55µs, db.query datasette-local 357µs, db.query.execute datasette-local 197µs, db.query datasette-local 330µs, db.query.execute datasette-local 73µs, db.query datasette-local 2.26ms, db.query.execute datasette-local 2.02ms, db.query datasette-local 232µs, db.query.execute datasette-local 64µs, db.query datasette-local 207µs, db.query.execute datasette-local 71µs, db.query datasette-local 6.11ms. The child span bars start progressively later across the early part of the timeline.

Introducing Mistral Large 4: Le chonk (via) Mistral are back in the game. Today they're releasing a preview of Mistral Large 4, a 1 trillion parameter, 49 billion active parameter model trained on their own cluster of 3,800 NVIDIA Grace Blackwell GPUs.

The preview is available via their API. They promise to release the open weights model at the "end of this month".

The model only supports two reasoning levels - "none" and "high" - via the Mistral API. Here are both pelicans - the "high" one looks better, though surprisingly it only used 2,717 output tokens compared to "none" which used 3,275:

It's good. The pouch is great, the bicycle frame is the right size, it has feet on pedals. Both pedals appear in front of the frame though. Nice gradients.

On Artificial Analysis it scores 38, just behind DeepSeek 4.1 Flash, which is a 552B model. It's a huge improvement on last December's Mistral Large 3, which drew this terrible pelican and scored 9 on AA.

It's certainly not a Fable-class model, but it's great to see Mistral put out a model that's back to being maybe about 6 months behind the frontier.

# 8:18 pm / ai, generative-ai, llms, mistral, pelican-riding-a-bicycle, llm-release

Comment My comment on EmbeddingGemma 2 — Hacker News

I really appreciate that EmbeddingGemma 2 is under the Apache 2.0 license.

For embedding models in particular, I don't think it makes sense to use a closed, proprietary, hosted-only model.

Most applications of embedding models involve calculating thousands or even millions of embedding vectors and storing them for later comparison.

If your model is proprietary, the vendor is likely someday going to decide to stop offering that model. They'll have a better model to replace it, but you still need to pay to re-calculate those millions of stored existing vectors.

(In April 2024 OpenAI offered to "cover the financial cost of users re-embedding content with these new models" - https://openai.com/index/gpt-4-api-general-availability/ - but I don't think that's something we can rely on from every provider.)

Notably, I don't want to host the model myself. I'd much rather pay a provider for a hosted model while knowing that if they ever stop hosting it I can run the open weights version myself - or find another vendor who can do that for me.

# 8:37 pm / google, ai, generative-ai, embeddings, gemma

Adds support for reasoning models, such as the newly released Mistral Large 4.

OpenAI released their new Jev-style Decisions API, as previously announced at last week's DevDay.

Since I already have an llm-typesafe plugin for talking to Jev, I had GPT-6 Astra read the new OpenAI API documentation and build an llm-openai-decisions plugin inspired by llm-typesafe.

Unlike Jev, the new gpt-6-luna decision model supports image input in addition to text. Both models charge for input it and not for output: OpenAI's is 10 cents per million input tokens, Jev's is 4.2 cents per million.

Otherwise the API shape is very similar to Jev, at least conceptually. Jev supports three question types for yes/no, choices, or scores. OpenAI Decisions supports the same three types.

Install the plugin like this:

llm install llm-openai-decisions

Here's an example query against an image attachment:

llm -m openai-decisions/gpt-6-luna \
  -a https://static.simonwillison.net/static/2025/two-pelicans.jpg \
  -s 'Does this image contain any mammals?'

And example output:

{"type": "predicate", "name": "evaluation", "probability": 0.0}

Consult the README for full details of how to run the other types of questions.

Since the Medicare breach, OpenAI has put in place additional monitoring to allow “immediate intervention” by staff to stop training if the company’s models access the internet in ways they’re not supposed to, Mr. Kwon [chief strategy officer at OpenAI] said.

— Victoria Kim, Reporting from the Australian parliament

# 11:58 pm / ai, openai, generative-ai, llms, ai-security-research, accidental-cyberattacks

← Monday, 5th October 2026
Wednesday, 7th October 2026 →

2026 » October

MTWTFSS
   1234
567891011
12131415161718
19202122232425
262728293031