BREAKING 🚨: ANTHROPIC ANNOUNCED CYBERSECURITY PROJECT GLASSWING AND MYTHOS BENCHMARKS!
Claude Mythos scored 93.9% on SWE Bench Verified and 87.3 on SWE Bench Multilingual!
“We do not plan to make Claude Mythos Preview generally available, but our eventual goal is to enable our users to safely deploy Mythos-class models at scale”
Claude Mythos scored 93.9% on SWE Bench Verified and 87.3 on SWE Bench Multilingual!
“We do not plan to make Claude Mythos Preview generally available, but our eventual goal is to enable our users to safely deploy Mythos-class models at scale”
❤5🔥5👍2
Anthropic announces Claude Mythos for cybersecurity research
Anthropic introduced Claude Mythos Preview, an AI model that autonomously detects and exploits zero-day vulnerabilities. It has uncovered critical flaws across major systems and is available to select partners, with $100 million in credits supporting cybersecurity efforts.
🗞 #claude
Anthropic introduced Claude Mythos Preview, an AI model that autonomously detects and exploits zero-day vulnerabilities. It has uncovered critical flaws across major systems and is available to select partners, with $100 million in credits supporting cybersecurity efforts.
🗞 #claude
TestingCatalog
Anthropic announces Claude Mythos for cybersecurity research
What's new? AnthropiC unveiled Claude Mythos Preview to spot zero-day flaws and craft exploits; select partners get access via Claude API for limited testing;
❤4👎1
Zhipu AI launches open-source GLM-5.1 model for coding tasks
Z AI has launched GLM-5.1, a flagship model built for agentic engineering and long-horizon coding, capable of running up to eight hours on a single task.
🗞 #ai
Z AI has launched GLM-5.1, a flagship model built for agentic engineering and long-horizon coding, capable of running up to eight hours on a single task.
🗞 #ai
TestingCatalog
Zhipu AI launches open-source GLM-5.1 model for coding tasks
GLM-5.1 by Z.ai debuts as a coding-focused AI model supporting long autonomous tasks, now available for all GLM Coding Plan users.
❤4👍3
Mythos grade intelligence might become available to users sooner than “months”.
> it’ll probably be months before we use a model of this level of capability
> Uhm
Soon 👀
> it’ll probably be months before we use a model of this level of capability
> Uhm
Soon 👀
xAI is training 7 different models on Colossus 2 in different sizes from 1T to 10T, including Imagine V2.
Not soon 👀
Not soon 👀
❤7👍2
BREAKING 🚨: Meta updated its Meta AI app with a slightly new design as well as its underlying model.
“I am Meta AI, powered by Muse Spark from the Muse model family.”
It constantly refers to the Muse model family and responses seem to be a bit different from earlier tested Avocado models.
Stealth launch 👀
“I am Meta AI, powered by Muse Spark from the Muse model family.”
It constantly refers to the Muse model family and responses seem to be a bit different from earlier tested Avocado models.
Stealth launch 👀
👀5👍4❤1
🚨 AI News | TestingCatalog
BREAKING 🚨: Meta updated its Meta AI app with a slightly new design as well as its underlying model. “I am Meta AI, powered by Muse Spark from the Muse model family.” It constantly refers to the Muse model family and responses seem to be a bit different…
Cyberpunk robot SVG from a newly released Muse Spark from Meta.
Not bad 👀
Not bad 👀
❤6👍3
🚨 AI News | TestingCatalog
BREAKING 🚨: Meta updated its Meta AI app with a slightly new design as well as its underlying model. “I am Meta AI, powered by Muse Spark from the Muse model family.” It constantly refers to the Muse model family and responses seem to be a bit different…
BREAKING 🚨: META ANNOUNCED MUSE SPARK, THE FIRST MSL MODEL, AND A NEW MUSE SPARK CONTEMPLATING MODE!
Muse Spark Contemplating mode scored 58.4% on HLE with tools!
"We’re also releasing Contemplating mode, which orchestrates multiple agents that reason in parallel. This allows Muse Spark to compete with the extreme reasoning modes of frontier models such as Gemini Deep Think and GPT Pro. Contemplating mode provides significant capability improvements in challenging tasks, achieving 58% in Humanity’s Last Exam and 38% in FrontierScience Research."
Muse Spark Contemplating mode scored 58.4% on HLE with tools!
"We’re also releasing Contemplating mode, which orchestrates multiple agents that reason in parallel. This allows Muse Spark to compete with the extreme reasoning modes of frontier models such as Gemini Deep Think and GPT Pro. Contemplating mode provides significant capability improvements in challenging tasks, achieving 58% in Humanity’s Last Exam and 38% in FrontierScience Research."
👍7❤3
🚨 AI News | TestingCatalog
BREAKING 🚨: META ANNOUNCED MUSE SPARK, THE FIRST MSL MODEL, AND A NEW MUSE SPARK CONTEMPLATING MODE! Muse Spark Contemplating mode scored 58.4% on HLE with tools! "We’re also releasing Contemplating mode, which orchestrates multiple agents that reason in…
Muse Spark will be available in private preview via API to select partners. Meta also "hopes" to open-source future versions of their models.
The model is already available to all users for testing on Meta AI.
The model is already available to all users for testing on Meta AI.
❤7👍4🔥3
🚨 AI News | TestingCatalog
BREAKING 🚨: META ANNOUNCED MUSE SPARK, THE FIRST MSL MODEL, AND A NEW MUSE SPARK CONTEMPLATING MODE! Muse Spark Contemplating mode scored 58.4% on HLE with tools! "We’re also releasing Contemplating mode, which orchestrates multiple agents that reason in…
Meta jumped from last to the 4th place on Artificial Analysis arena with its newly released Muse Spark model.
It also appears to be token efficient for its level of intelligence.
It also appears to be token efficient for its level of intelligence.
❤11🔥4👍3
This media is not supported in your browser
VIEW IN TELEGRAM
BREAKING 🚨: Anthropic announced Claude Managed Agents in public beta on Claude Platform!
Claude Managed Agents will allow businesses to deploy agents with the latest capabilities at scale!
Conway, phase 1👀
Claude Managed Agents will allow businesses to deploy agents with the latest capabilities at scale!
Conway, phase 1
Please open Telegram to view this post
VIEW IN TELEGRAM
❤3
Media is too big
VIEW IN TELEGRAM
ICYMI: Google is hosting Google Cloud Next 26 event on April 22-24 in Las Vegas.
Loads of updates are expected across Gemini, Vertex AI, Agents, Vibe Coding, Generative UI, and AI Cloud.
> Going to be a fun next few months 👀
Loads of updates are expected across Gemini, Vertex AI, Agents, Vibe Coding, Generative UI, and AI Cloud.
> Going to be a fun next few months 👀
❤5🔥2
Perplexity will be running an 8-weeks competition where contenders will use Perplexity Computer to build businesses.
Up to 1M investment and up to 1M in Computer credits.
We are slowly shifting towards levels 4 (innovators) and 5 (AI orgs) on the path to AGI and more and more AI solutions for businesses will arrive during 2026.
Up to 1M investment and up to 1M in Computer credits.
We are slowly shifting towards levels 4 (innovators) and 5 (AI orgs) on the path to AGI and more and more AI solutions for businesses will arrive during 2026.
❤4🔥3
Meta unveils Muse Spark AI model and new Contemplating mode
Meta has launched Muse Spark, a new AI model built on a rebuilt pretraining stack with upgraded architecture, optimization, and data pipelines. It matches prior performance using over 10x less compute and ranks 4th on Artificial Analysis, targeting scalable research use.
🗞 #meta
Meta has launched Muse Spark, a new AI model built on a rebuilt pretraining stack with upgraded architecture, optimization, and data pipelines. It matches prior performance using over 10x less compute and ranks 4th on Artificial Analysis, targeting scalable research use.
🗞 #meta
TestingCatalog
Meta unveils Muse Spark AI model and new Contemplating mode
What's new? Muse Spark is a new AI model scaling toward personal superintelligence with refined pretraining; benchmarks show tenfold lower compute use;
❤3👍2
Anthropic launches Claude Managed Agents for businesses
Anthropic has launched Claude Managed Agents in public beta, offering composable APIs and production-grade infrastructure for building and deploying cloud-hosted AI agents. The suite supports secure execution, orchestration, and multi-agent workflows at scale.
🗞 #claude
Anthropic has launched Claude Managed Agents in public beta, offering composable APIs and production-grade infrastructure for building and deploying cloud-hosted AI agents. The suite supports secure execution, orchestration, and multi-agent workflows at scale.
🗞 #claude
TestingCatalog
Anthropic launches Claude Managed Agents for businesses
What's new? Anthropic released Claude Managed Agents, a suite of apis for cloud-hosted AI agents with secure code execution and session tracing in public beta;
❤4
This media is not supported in your browser
VIEW IN TELEGRAM
BREAKING 🚨: Google has integrated NotebookLM directly into Gemini!
It will enable users to work with notebooks directly in the Gemini UI and use Gemini chats as sources for NotebookLM.
"We're rolling out notebooks in Gemini today, starting with Google AI Ultra, Pro, and Plus subscribers on the web. In the coming weeks, we'll expand access to mobile, more countries across Europe, and to free users."
It will enable users to work with notebooks directly in the Gemini UI and use Gemini chats as sources for NotebookLM.
"We're rolling out notebooks in Gemini today, starting with Google AI Ultra, Pro, and Plus subscribers on the web. In the coming weeks, we'll expand access to mobile, more countries across Europe, and to free users."
❤12🔥4
Meta stock price went up by 6-8% after Muse Spark model announcement.
👍6❤4