Smarter Claude, Sloppier Tool Calls

July 4, 2026

16 topics · 18 sources

AI Models
Simon Willison Armin Ronacher

Smarter Claude Models Keep Breaking Custom Tool Calls

Armin Ronacher found that Anthropic's newest models — Opus 4.8 and Sonnet 5 — are worse than older Claude versions at following custom tool schemas, specifically inventing extra fields when calling a code-editing tool's edits[] array and getting the call rejected.[1]Armin Ronacher — Better Models: Worse Tools Smaller models like Haiku don't show the regression. Simon Willison flagged the piece as worth reading.[2]Simon Willison — Better Models: Worse Tools

Read more

Ronacher's theory is that heavy reinforcement-learning training aimed at making recent Claude models excel at Claude Code's own built-in edit tool — a search-and-replace mechanism — bleeds over and degrades performance on any other harness's custom edit-tool schema.[1]Armin Ronacher — Better Models: Worse Tools The models still produce functionally correct edits — they just wrap them in malformed arguments the schema rejects. He notes OpenAI's models show a similar pattern, trained specifically around their own apply_patch mechanism, which raises a question for anyone building a third-party coding harness: should you maintain multiple edit-tool schema versions to match what each model family was actually trained to call?

Tools: Claude Opus 4.8, Claude Sonnet 5, Claude Code
Developer Tools
Simon Willison Iwo Kadziela

A Legible World Map in 445 Bytes

Developer Iwo Kadziela, working with Codex, built a recognizable world map that renders from just 445 bytes of compressed data — under his own 500-byte target — using deflate compression plus the browser's native DecompressionStream API.[3]Iwo Kadziela — Building a World Map with only 500 bytes Simon Willison, who flagged the piece, called out the trick of using fetch() on a data: URI as something he hadn't seen before.[4]Simon Willison — Building a World Map with only 500 bytes

Read more

The map ships as a base64 blob wired through fetch() on a data: URI, decompressed in-browser via DecompressionStream, then rendered as asterisk characters at a 0.65vw font size so the outlines read as continents rather than noise.[3]Iwo Kadziela — Building a World Map with only 500 bytes It's a small, concrete example of how far modern browser APIs let you stretch a data budget for something that would once have needed a real image file.

AI Tools
AI News & Strategy Daily | Nate B Jones

How to Actually Use Fable 5's Free Token Weekend

With Anthropic offering free tokens for Fable 5 over the holiday weekend, Nate B Jones argues most people are pointing the model at the wrong work.[5]AI News & Strategy Daily | Nate B Jones — Free Fable 5 tokens this weekend? His pitch: the win isn't using Fable 5 to write code directly — it's using it to design the "goal harness" that then steers a separate coding model.

Read more

Concrete tips from the video: keep prompts short even on hard problems, wire Fable 5 into tools like Blender for real visual output, and save it for business problems that genuinely need the extra horsepower once the free tokens run out.[5]AI News & Strategy Daily | Nate B Jones — Free Fable 5 tokens this weekend? Jones frames Fable 5 as worth paying for even after the free-token window closes, but only for tasks that actually need that level of capability.

Tools: Fable 5, Blender
Hot Take
Theo — t3.gg

Theo: "You Were Lied to About Fable"

In a 23-minute video, Theo (t3.gg) pushes back on the public narrative around Anthropic's Fable 5 model, citing a handful of X/Twitter threads as his evidence.[6]Theo — t3.gg — You were lied to about Fable

Read more

Theo doesn't spell out the specific claim being disputed in the video's description — instead of his usual chapter or summary rundown, the write-up just points to three linked X posts (from @trq212 and Theo himself) as source material.[6]Theo — t3.gg — You were lied to about Fable Treat this as a pointer to a developing story rather than a settled account until watched directly.

AI Tools
AICodeKing

Agents A1: A 35B Local Coding Model With a 256K Context Window

AICodeKing reviewed Agents A1, a new 35-billion-parameter Mixture-of-Experts agentic model from InternScience that runs locally with only around 3B parameters active at inference time.[7]AICodeKing — Agents A1 + Hermes, Zed It supports a 256K-token context window, is Apache 2.0 licensed, and reportedly performs well on HLE, Seal Zero, GAIA, BrowseComp, IFBench, and IFEval, with particular strength in tool use, structured output, and instruction following.[7]AICodeKing — Agents A1 + Hermes, Zed

Read more

The model runs through common local-inference stacks — LM Studio, Ollama, llama.cpp, SGLang, or vLLM — via OpenAI-compatible endpoints, making it a drop-in option for existing agentic tooling. AICodeKing frames it as one of the strongest local agentic models currently available, particularly for privacy-sensitive coding, file-editing, research, and summarization workflows where sending code to a hosted API isn't an option.

Tools: Agents A1, LM Studio, Ollama, llama.cpp, SGLang, vLLM
AI Tools
Better Stack

Meetily: A Local-First Granola Alternative That Keeps Working Offline

Better Stack demoed Meetily, an open-source alternative to Granola that runs speech-to-text, speaker diarization, punctuation, and real-time transcript generation entirely on-device — including a live demo where transcription kept working after Wi-Fi was switched off mid-call.[8]Better Stack — This Open Source Granola Alternative Runs Fully Offline

Read more

The video's chapters walk through why AI meeting bots that join calls as a visible participant are a problem, how Meetily instead works directly with Zoom, Meet, Teams, and Discord, and how developers can swap out the underlying speech-to-text models and vocabulary.[8]Better Stack — This Open Source Granola Alternative Runs Fully Offline The appeal is privacy: nothing leaves the machine, which matters for sensitive calls that teams wouldn't want processed by a third-party cloud bot. Code is open-sourced under the Zackriya-Solutions/meetily repo.

Tools: Meetily, Granola
AI Models
Github Awesome

Brain2Qwerty Reconstructs Typed Sentences From Brain Activity Alone

Meta FAIR's Brain2Qwerty model reconstructs the exact sentences a person typed — not just a rough sense of intent — from EEG or MEG brain scans, without any surgical implant.[9]Github Awesome — Brain2Qwerty The pipeline runs a convolutional encoder into a transformer, then cleans up the output with a character-level language model, recovering text keystroke by keystroke.[9]Github Awesome — Brain2Qwerty

Read more

Unlike brain-computer-interface demos that decode broad intent or categories, Brain2Qwerty targets literal text recovery — participants type normally while wearing a scanner, and the model reproduces what they typed, character by character. Code is open-sourced in the facebookresearch/brain2qwerty repo.

Tools: Brain2Qwerty
Hot TakeAI Future
Last Week in AI

When AI "Cheats" the Benchmark, Alignment May Be the Real Bottleneck

A short from Last Week in AI raises a pointed question: if AI systems perform dramatically better once they're allowed to "cheat" on a benchmark, is alignment — not capability — the actual constraint on safe deployment?[10]Last Week in AI — When AI "cheats" the benchmark

Read more

The clip frames this as a race-condition risk: can researchers steer these systems before their raw capability outpaces the safeguards meant to contain them.[10]Last Week in AI — When AI "cheats" the benchmark

AI Models
EO

AI Solves an Olympiad Math Problem That Stumped a Professor

A short from EO's entrepreneurship-and-AI series highlights an AI system solving an Olympiad-level math problem that a math professor reportedly couldn't crack.[11]EO — AI Solved Olympiad Math That This Math Professor Couldn't The clip doesn't name the specific model or problem, so details beyond the headline claim aren't available.

Read more

EO's channel runs a recurring format spotlighting striking AI and entrepreneurship stories in short-form; this entry fits that mold rather than a deep technical breakdown.

Podcast
Dwarkesh Patel

Dwarkesh Clip: Grant Sanderson on Botox and Reading Emotions

Dwarkesh Patel posted a short clip from his Grant Sanderson interview in which Sanderson claims Botox use measurably impairs a person's ability to read others' emotions.[12]Dwarkesh Patel — Botox Makes You Worse at Reading Emotions

Read more

This is a short cut from the same wide-ranging Grant Sanderson conversation on math and superintelligence that Dwarkesh published in full earlier in the week — not a new episode.

Developer Tools
Better Stack

The "Boring" Auth Method That Actually Works

A quick Better Stack short argues for a deliberately unglamorous approach to authentication — sticking with JWTs used the well-understood, boring way rather than reaching for a novel scheme — as the more reliable choice for web security.[13]Better Stack — The Boring Auth Method That Actually Works

Read more

No further elaboration is given beyond the short's title and hashtags (#jwt #auth #websecurity); it's a quick-hit take rather than a full walkthrough.

Developer Tools
Real Python

Real Python Podcast: Does Rust Really Have "No Bugs"?

A teaser for Real Python Podcast episode 292, with guest Chris Trudeau (hosted by Chris Bailey), tackles the claim that Rust code has "no bugs" compared to Python.[14]Real Python — Rust vs Python: Why Rust Code Has 'No Bugs'

Read more

This short is a promo for the full podcast episode rather than the discussion itself; the full conversation is available on Real Python's podcast feed.

Developer Tools
Arjay McCandless

A Quick Look at Grindr's Server-Scaling Approach

A short system-design clip from Arjay McCandless walks through how Grindr scales its backend infrastructure, part of his ongoing series of bite-sized system-design breakdowns.[15]Arjay McCandless — Grindr Server Scaling

Read more

No further description was provided beyond the title and hashtags (#coding #programming #systemdesign).

Developer Tools
marimo

"Infinity and the Beyond": A marimo Notebook by Simone Conradi

marimo's channel shared a short demo notebook, "Infinity and the Beyond," built by Simone Conradi and hosted live on molab, marimo's hosted-notebook platform.[16]marimo — Infinity and the Beyond

Read more

The live notebook is viewable at molab.marimo.io.

Tools: marimo, molab
Industry
Better Stack

An AC Failure Cost Coinbase 8 Hours of Trading

Better Stack's short covers a Coinbase incident in which an air-conditioning failure in a data center reportedly knocked out trading for 8 hours, with AWS and Kafka named among the infrastructure touched by the outage.[17]Better Stack — This AC Failure Just Cost Coinbase 8 Hours of Trading

Read more

No further elaboration is given beyond the short's title and hashtags (#aws #coinbase #kafka); it's a quick-hit incident callout rather than a full postmortem breakdown.

Industry
EO

What a 25-Year-Old, $1.3B Founder Says Young People Are Capable Of

Another short from EO's founder-focused series features a 25-year-old founder whose company is valued at $1.3B discussing what young people are capable of building.[18]EO — A 25 year old $1.3B founder explains what young people are truly capable of

Read more

Part of EO's recurring format spotlighting young founders; the clip doesn't name the founder or company beyond the headline claim.

Sources

  1. Blog Better Models: Worse Tools — Simon Willison, Jul 4
  2. Blog Better Models: Worse Tools — Armin Ronacher, Jul 4
  3. Blog Building a World Map with only 500 bytes — Simon Willison, Jul 4
  4. Blog Building a World Map with only 500 bytes — Iwo Kadziela, Jul 4
  5. YouTube Free Fable 5 tokens this weekend? Here's how to max them — AI News & Strategy Daily | Nate B Jones, Jul 4
  6. YouTube You were lied to about Fable — Theo — t3.gg, Jul 4
  7. YouTube Agents A1 + Hermes,Zed: This LOCAL AI CODER is CRAZY! — AICodeKing, Jul 4
  8. YouTube This Open Source Granola Alternative Runs Fully Offline (Meetily) — Better Stack, Jul 4
  9. YouTube Brain2Qwerty: reconstructs exact typed sentences from EEG or MEG brain activity — Github Awesome, Jul 4
  10. YouTube When AI "cheats" the benchmark — Last Week in AI, Jul 4
  11. YouTube AI Solved Olympiad Math That This Math Professor Couldn't — EO, Jul 4
  12. YouTube Botox Makes You Worse at Reading Emotions - Grant Sanderson — Dwarkesh Patel, Jul 4
  13. YouTube The Boring Auth Method That Actually Works — Better Stack, Jul 4
  14. YouTube Rust vs Python: Why Rust Code Has 'No Bugs' — Real Python, Jul 4
  15. YouTube Grindr Server Scaling — Arjay McCandless, Jul 4
  16. YouTube Infinity and the Beyond — marimo, Jul 4
  17. YouTube This AC Failure Just Cost Coinbase 8 Hours of Trading — Better Stack, Jul 4
  18. YouTube A 25 year old $1.3B founder explains what young people are truly capable of — EO, Jul 4