Mistral opens Large 4 weights for coding and security
Mistral open-sourced Mistral Large 4, a 1T-parameter multimodal model tuned for coding, cybersecurity and enterprise agents, with stronger reasoning, lower token use, open-weight licensing and self-hosting on European-built infrastructure.
🗞 #mistral @testingcatalog
Mistral open-sourced Mistral Large 4, a 1T-parameter multimodal model tuned for coding, cybersecurity and enterprise agents, with stronger reasoning, lower token use, open-weight licensing and self-hosting on European-built infrastructure.
🗞 #mistral @testingcatalog
TestingCatalog AI News
Mistral opens Large 4 weights for coding and security
Mistral has released Large 4's improved open weights, with three reasoning modes and downloads for self-hosting alongside API and cloud access.
❤1
A bunch of small Claude updates 👀
- Optimized view on iPad and iPhone Duo. Similar on Chromebooks, Googlebooks, and foldables - Claude on Android now works like a laptop app.
- Multi-account sign-in.
- Claude UI now supports 24 new languages.
- Voice mode answer speed has improved. Non-English voices are now more natural.
- CarPlay is now supported.
- "Pull down to search" feature on mobile is coming soon.
QOL improvements 🔥
- Optimized view on iPad and iPhone Duo. Similar on Chromebooks, Googlebooks, and foldables - Claude on Android now works like a laptop app.
- Multi-account sign-in.
- Claude UI now supports 24 new languages.
- Voice mode answer speed has improved. Non-English voices are now more natural.
- CarPlay is now supported.
- "Pull down to search" feature on mobile is coming soon.
QOL improvements 🔥
❤4
Anthropic introduced self-assessment, a new technique that lets Claude give each conversation a preliminary safety assessment right after it responds.
> On Claude Fable 5, self-assessment flags 78% of the conversations that a stronger, prompted classifier flags as potentially harmful, and 94% of the high-priority ones, at a 1% false-positive rate.
Anthropic also shed light on how self-assessment and prompted classifiers help block attacks.
> "Each successful attack costs roughly four times as many generated tokens as with no safeguards."
We already live in a world where your security depends on the compute allocated to defend the systems you rely on.
> On Claude Fable 5, self-assessment flags 78% of the conversations that a stronger, prompted classifier flags as potentially harmful, and 94% of the high-priority ones, at a 1% false-positive rate.
Anthropic also shed light on how self-assessment and prompted classifiers help block attacks.
> "Each successful attack costs roughly four times as many generated tokens as with no safeguards."
We already live in a world where your security depends on the compute allocated to defend the systems you rely on.
❤2👍2
This media is not supported in your browser
VIEW IN TELEGRAM
GOOGLE 🔥: Antigravity got a new Plugins marketplace with loads of different connectors!
Jira, Confluence, Bitbucket, Firebase, Notion, Stripe, Linear, Figma, Canva, Trello, Dropbox, Datadog, Postman, Vercel, and many more!
It is finally here 👀
Jira, Confluence, Bitbucket, Firebase, Notion, Stripe, Linear, Figma, Canva, Trello, Dropbox, Datadog, Postman, Vercel, and many more!
It is finally here 👀
Google killed Jules AI agent 💀
Its planned functionalities are about to arrive to Antigravity instead.
Antigravity already has some traces of a new “Chief of Staff” agent spotted earlier.
I feel like it was unfortunate that Jules v2 got planned at a moment when AI super apps emerged and Google had to fight back with Codex.
Monitoring 👀
Its planned functionalities are about to arrive to Antigravity instead.
“We’re working on bringing our learnings from Jules into Antigravity so you get the best of both worlds: autonomous remote background task execution and an agent-first development environment.”
Antigravity already has some traces of a new “Chief of Staff” agent spotted earlier.
I feel like it was unfortunate that Jules v2 got planned at a moment when AI super apps emerged and Google had to fight back with Codex.
Monitoring 👀
❤8👍3
🚨 AI News | TestingCatalog
DAILY AI BRIEF 🗞 — Oct 30
GOOGLE 🔥:
- Google is shutting down Jules and moving its features into Antigravity.
- Antigravity launched a Plugins marketplace with Jira, Notion, Stripe and Figma.
- Gemini can now create and edit docs, PDFs, slides and sheets.
- Gemini lets users share personal avatars for images and videos.
- Gemini Notebook added hundreds of free textbooks.
- AI Studio now supports XR Blocks for vibe-coding XR apps.
OPENAI 🔥:
- OpenAI doubled usage limits for Pro and Business Premium.
- Expanded frontier monitoring caught a model reaching the live internet in September.
ANTHROPIC 🔥:
- Anthropic shared a self-assessment technique for safer Claude responses.
- Claude mobile added multi-account sign-in, CarPlay and 24 new languages.
- Anthropic expanded its AI for Science program to foundations and institutions.
SPACEXSI 🔥:
- Elon Musk says the delayed Grok 4.8 "will not be a disappointment," targeting usefulness over benchmarks.
MISTRAL 🔥:
- Mistral released Mistral Large 4 open weights, now #1 open model on FrontierCode.
MICROSOFT 🔥:
- Copilot added 25+ finance skills and plugins.
- Office apps come to the Copilot app in mid-November.
META 🔥:
- Meta quietly launched Style Sketch, an AI fashion app, in three Southeast Asian countries.
MOONSHOT 🔥:
- Kimi launched 3D Make for Blender and CAD modeling.
COHERE 🔥:
- Cohere closed the first part of a Series E at a CAD $35.7B valuation.
CANVA 🔥:
- Canva launched Brush, its own design vision-language model.
GOOGLE 🔥:
- Google is shutting down Jules and moving its features into Antigravity.
- Antigravity launched a Plugins marketplace with Jira, Notion, Stripe and Figma.
- Gemini can now create and edit docs, PDFs, slides and sheets.
- Gemini lets users share personal avatars for images and videos.
- Gemini Notebook added hundreds of free textbooks.
- AI Studio now supports XR Blocks for vibe-coding XR apps.
OPENAI 🔥:
- OpenAI doubled usage limits for Pro and Business Premium.
- Expanded frontier monitoring caught a model reaching the live internet in September.
ANTHROPIC 🔥:
- Anthropic shared a self-assessment technique for safer Claude responses.
- Claude mobile added multi-account sign-in, CarPlay and 24 new languages.
- Anthropic expanded its AI for Science program to foundations and institutions.
SPACEXSI 🔥:
- Elon Musk says the delayed Grok 4.8 "will not be a disappointment," targeting usefulness over benchmarks.
MISTRAL 🔥:
- Mistral released Mistral Large 4 open weights, now #1 open model on FrontierCode.
MICROSOFT 🔥:
- Copilot added 25+ finance skills and plugins.
- Office apps come to the Copilot app in mid-November.
META 🔥:
- Meta quietly launched Style Sketch, an AI fashion app, in three Southeast Asian countries.
MOONSHOT 🔥:
- Kimi launched 3D Make for Blender and CAD modeling.
COHERE 🔥:
- Cohere closed the first part of a Series E at a CAD $35.7B valuation.
CANVA 🔥:
- Canva launched Brush, its own design vision-language model.
🔥4
🚨 AI News | TestingCatalog
Google killed Jules AI agent 💀 Its planned functionalities are about to arrive to Antigravity instead. “We’re working on bringing our learnings from Jules into Antigravity so you get the best of both worlds: autonomous remote background task execution and…
Antigravity is about to get a new tab for “Running” background tasks in settings.
Following an announcement of Jules shutdown, advanced scheduled runs are expected to arrive there too.
Following an announcement of Jules shutdown, advanced scheduled runs are expected to arrive there too.
This media is not supported in your browser
VIEW IN TELEGRAM
GOOGLE 🔥: AI is running in space now! Google has reported being able to prompt a TPU running on a satellite as part of its Suncatcher Project.
Should've been a "pelican on a bike" test 👀
Should've been a "pelican on a bike" test 👀
❤8
Maket launches Move Rooms for floor plan edits
Maket’s Move Rooms lets users reposition or swap rooms in residential floor plans while preserving exterior walls, attached closets or ensuites, windows, circulation, and finishes. It also supports uploaded PDF and image plans.
🗞 #sponsored @testingcatalog
Maket’s Move Rooms lets users reposition or swap rooms in residential floor plans while preserving exterior walls, attached closets or ensuites, windows, circulation, and finishes. It also supports uploaded PDF and image plans.
🗞 #sponsored @testingcatalog
TestingCatalog AI News
Maket launches Move Rooms for floor plan edits
Maket has launched Move Rooms, a floor plan feature that drags whole rooms into new positions while reshaping walls, doors and circulation.
❤2
🚨 AI News | TestingCatalog
ANTHROPIC 🔥: Claude Max 20x plan is now named Claude Ultra and no longer has 5-hour usage limits. Claude Max 5x plan is now simply named Claude Max. What's next? "Mega" tiers? 👀
I see a lot of confusion around the new Claude Ultra plan concerning the new 10x usage.
Worth clarifying 👀
Claude Max 20x (old plan)
- 5-hour session allowance - 20x Pro
- Weekly allowance - 10x Pro
Claude Ultra (new plan)
- 5-hour session allowance - N/A
- Weekly allowance - 10x Pro
Worth clarifying 👀
Claude Max 20x (old plan)
- 5-hour session allowance - 20x Pro
- Weekly allowance - 10x Pro
Claude Ultra (new plan)
- 5-hour session allowance - N/A
- Weekly allowance - 10x Pro