AI posts
People and companies shaping AI. Filter your watched accounts, catch up on recent posts, or find the most liked and viewed.
Compiled by the Toolstash team · Updated
209 watched accounts · 2978 matching posts · Showing up to 3,000 recent stored posts. Counts are snapshots, not live X totals. This page refreshes every minute while visible. Last collected .
@MaziyarPanahi ·
i gave Baidu's new open OCR model a doctor's drug manual from 1899. all 270 pages in one request, on one GPU
left is the scan, right is what it read, in its own layout
the cocaine entry lists morphine as an antidote
what's the oldest paper record your hospital keeps?
@DeryaTR_ ·
Why I believe AI / SI will unlock cures for every Cancer within the next five to ten years ears - PART 1:
For decades, cancer researchers have confronted a seemingly impossible problem: the biological complexity of cancer is so enormous that no individual scientist, research team, or even the entire scientific community can systematically explore all possible therapeutic solutions.
Consider what happens inside a single tumor. Thousands of genes, proteins, metabolites, and signaling pathways interact across millions or billions of cells. Then cancer cells constantly evolve, acquiring mutations, adapting to their environments, escaping immune surveillance, and developing resistance to treatments. Their behavior is further influenced by surrounding immune cells, blood vessels, fibroblasts, and the metabolic conditions within the tumor. And every patient's cancer is different.
This creates an astronomical number of possible biological states and therapeutic interventions. Even when scientists identify a promising molecular target, they must determine whether interfering with it will actually kill the cancer, whether healthy tissues depend on the same mechanism, and whether the tumor will eventually escape.
Historically, we’ve approached this problem through relatively very slow cycles of hypothesis generation, experimentation, and clinical testing. We investigate a limited number of targets, screen a fraction of possible compounds, and spend years determining whether a treatment works.
Now imagine what happens when the intelligence available to solve these problems increases exponentially.
I believe the next five to ten years will bring superintelligence systems whose scientific reasoning, computational modeling, and experimental capabilities will vastly exceed those of today's best researchers put together!
The implications for cancer research will be profound. Here is the part 1 of how this can be achieved:
Virtual cells: One of the most exciting developments will be the emergence of virtual cells, or computational models designed to predict how living cells behave and respond to interventions
Imagine an AI system that can represent a cancer cell's molecular machinery, including its genetic mutations, gene expression, protein interactions, metabolism, and signaling networks. Instead of testing one drug at a time, scientists could ask the model what would happen if they inhibited a particular protein, activated a specific immune pathway, edited a gene, or administered several treatments simultaneously.
Now extend that capability to millions of virtual cancer cells representing different tumor subpopulations, together with immune cells and healthy tissues. We could potentially simulate how an entire tumor responds to treatment, how resistant populations emerge, and how different interventions affect normal cells!
This is the vision of increasingly comprehensive biological foundation models: AI systems trained on extensive molecular and experimental data that learn generalizable principles of cellular behavior.
The extraordinary opportunity is that such systems could eventually discover therapeutic mechanisms that we have never considered because we lack that computational capability. They might identify previously unrecognized combinations of molecular vulnerabilities, predict how to overcome resistance, and design interventions that exploit subtle differences between malignant and healthy cells.
Today's virtual-cell models are still very limited approximations of biology. But their predictive accuracy and scope will improve dramatically, and they will become powerful engines, accelerating precise therapeutic discoveries by magnitudes faster!
Part 2 will be about the digital twins. To be continued.
@hwchase17 ·
crons are a simple technology but can be used in a bunch of ways to unlock new capabilities in agent harnesses: - memory (dreaming) - proactiveness - user reminders
and more!
@thdxr ·
we have an internal alternative to slack huddles with higher quality video, audio, screenshare
we will release it publicly but some weird ux issues to fix first...
@sophiamyang ·
Livestream in 40min. Join me to find out: - what's the secret sauce behind M3.1 Flash to make it fast and cheap? - does that low price turn into real useful work?
@ArtificialAnlys ·
GitHub has released Copilot Fusion in GitHub Copilot, which orchestrates between Anthropic and OpenAI models. It scores 2 points below Opus 5.5 in Claude Code at half the Cost per Task in the Coding Agent Index
GitHub describes Copilot Fusion as a runtime orchestration system, which selects and coordinates models to optimize performance, cost, and latency for each task. Copilot Fusion chooses between a number of models from Anthropic and OpenAI, including Claude Opus 5.5, Claude Sonnet 5.5, GPT-6 Astra, and GPT-6 Sol. We benchmarked its max setting in GitHub Copilot CLI in the Artificial Analysis Coding Agent Index v2.0.
Token costs are charged based on the underlying models used to complete a request.
Key takeaways:
➤ Cost-efficient for near-frontier performance: Copilot Fusion in GitHub Copilot costs $5.59 per task, versus $11.42 for Claude Code with Opus 5.5 (max). Devin Fusion with Opus 5.5 (max) + GPT-6 Luna (high) scores 62 at $8.00 per task.
➤ Slight lead in Terminal-Bench 4.0: The Coding Agent Index v2.0 weights its three constituent benchmarks equally. Copilot Fusion in GitHub Copilot leads Terminal-Bench 4.0 with a score of 71. It ranks third on FrontierCode in the chart at 51, behind Claude Code with Opus 5.5 (max) at 56 and Devin Fusion with Opus 5.5 (max) + GPT-6 Luna (high) at 54. Its SWE-Atlas-QnA-AA score lags at 57, between GPT-6 Astra in Codex and Gemini 4 Argon in Antigravity CLI.
➤ Half the token use of Opus 5.5 in Claude Code: GitHub Copilot Fusion used 7.2M tokens per task in the Coding Agent Index, approximately half of the tokens used by Claude Opus 5.5 in Claude Code, and 30% less than Devin Fusion with Opus 5.5 (max) + GPT-6 Luna (high).
@mattpocockuk ·
Looks like /drill-me is back on the menu
@dani_avila7 ·
Receipts were easy for MiniCPM-OCR by @OpenBMB
So I gave it the hard stuff: a 9-line invoice, a bank statement, a hand-corrected PO and a German–English hotel bill
82/102 fields exactly right. Headers and printed totals: 73/78. 🧵👇
@thorwebdev ·
Genius 👷🏗️🤯 both from Manus and Natalie 💪
Hope we'll get to see more of these micro startups within startups! 🔥
@AravSrinivas ·
Rolling out new voice mode for all Pro and Max users, on web and Mac apps. Allows you to basically talk to your computer to get parallel work streams done.
@FireworksAI_HQ ·
How fast is MiniMax M3.1 Flash?
@VictorSuOrtiz of @MiniMax_AI and Fireworks' @sophiamyang are going live this morning at 10 AM PT to break down what's new, what it's built for, and how to use it on Fireworks today.
Tune into the livestream here: https://x.com/i/broadcasts/2JQRWzVmqdQEVy
@thdxr ·
we have cloudflare enterprise now
which means we're paying for all your cheap worker deployments on all your side projects
you're welcome
@AndrewCurran_ ·
Deep expertise in any STEM field will increasingly become an infohazard over the next few years. Watching my friends in the field process the ETH result really cemented this. During this Weltbilderschütterung, be there for people who thought they built their houses upon stone.
@FireworksAI_HQ ·
We’re working to build the largest collection of private and public practitioner evals.
Today, 31 new partners join the Specialized Intelligence Index. Compare open, closed, and specialized models on real work, weigh quality against cost, and see where training could pay off.
@thsottiaux ·
Pretty wild that this site got 3M views.
@elonmusk ·
They do
@tibo_maker ·
wild what people say about it 🤯
@omarsar0 ·
Super interesting way to shop with AI.
Dalet hyper-personalizes the shopping experience by creating a feed of lifelike AI-generated images.
The products in each image are real. And users can tap to buy them.
Discovery is a way better fit for generative media than another chat box.
The team comes from Google, Runway, and MIT. Congrats on their raise.
@ManusAI ·
Manus didn't just build the site. it built the whole company.
the interactive site. the 3D brick builds. piece-by-piece build checks. Stripe. domain. email. launch video. social posts. even the ads.
Check out https://bricklane.online!
@paolino ·
DJs, stop preparing your tracks.
I built Solco: it grids, keys and tags your whole library on your computer, then writes the USB your CDJs read. No rekordbox needed.
12 local AI models, macOS, Windows, and Linux.
Join the waitlist: https://getsolco.com
@claudeai ·
New in Claude for Legal: a gallery of ready-made artifacts for contract review, regulatory horizon scans, cite checks, and more.
Start from the built-in example, add your own documents and connectors, and share it with anyone across the business.
@swyx ·
Happy anniversary to Ligma/Johnson 2026
@ben_burtenshaw ·
reading up on swarms and multi-agent systems a lot these days and trying to understand how you evaluate, how you reward.
as a default data person, my instinct is to scope out interpretable logical verifiers that work to check for the kind of collaborative behaviour you’d expect.
starting to realise this is a mistake. this paper avalon m1 from @GenReasoning , has a very simple and therefore probably good way of collectively rewarding all agents in the system for success.
- every agent gets the same team-level advantage, with no per-agent credit assignment.
- a ceo decides headcount, roles and tools before any work starts.
- agents generalize into ‘lead engineers’ that become hubs, and peer review
@arena ·
Where do users actually feel the difference with a frontier coding agent?
Across 20,869 Agent Arena traces, we compared how users criticized the five top models (all from @AnthropicAI & @OpenAI) on Agent Arena: Code, versus how they criticized other agents.
Here are 3 areas where frontier models make the most difference:
1. Less getting stuck: Among sessions with complaints, failed repair loops come up 6.9 percentage points less often, and project context gets ignored less often too (-1.9 pp). You spend less time steering the agent out of dead ends.
2. Better taste and code that works: They are better at design (-6.4 pp) and at producing working results (-5.8 pp). What you get back is more likely to be what you wanted.
3. More diligence: They follow instructions better (-5.6 pp), finish the work more often (-4.9 pp) and break existing work less (-4.5 pp). There's less to clean up afterwards.
Curious for a deeper dive into Agent Arena's praise vs complaint signal? See the article below in thread.
@svpino ·
It's interesting to think about how little baggage people who've never coded before bring to this agentic era.
If you haven't coded before, you can't waste time doing manual code reviews, so you need to use agents to review what you ship.
Your disadvantage becomes sort of an advantage that forces you to automate the process.
It takes mountains for people who did the work before to let go of all methods and embrace the new ways.
@dani_avila7 ·
Today is my last day in Florida after a few days at Disney and Universal with my family.
And I’m pretty proud of how disconnected I managed to stay.
Before leaving, I set up my AI agents to keep working and send me daily summaries.
@Muse was my main orchestrator.
It monitored my social accounts, spotted trends, blocked bots, and kept me updated without constantly checking my phone.
I also connected it to my email and Todoist. Anything Muse could handle, it did. Everything else became a task for when I got back.
Now I’m heading home without hundreds of emails and notifications to catch up on.
Thanks, Muse 👏👏
Probably my favorite use case for AI agents so far… actually being able to disconnect and enjoy time with my family.
@edwinarbus ·
I refuse to use an iPhone Duo on the train
I do not want people to look at me
Don’t look at me
@omarsar0 ·
Great upgrade to Drex Max.
Drex Max 1.1 is Nace's new decision model, with better out-of-distribution generalization. It's multimodal and has a 262K context window.
I've been testing Drex Max on research tasks, so I am curious to test out its better generalization to new tasks.
Nace reports 70.8 on the Decision Index (up from 67.6 for Drex Max) and a median latency of 172 ms.
@rauchg ·
fx is now much faster, smaller and efficient Try https://fx.sh on your iPhone Duo
@svpino ·
This is an interesting idea that replaces scrolling through Pinterest or Instagram when you are looking for inspiration:
Let's say you are looking for ways to decorate your home.
1. This will show you AI-generated images with real products: a sofa, a chair, lamps, a TV, etc.
2. The products will be part of a realistic-looking scene so you can see how they work together.
3. You can interact with the different products, and this will shape what the app shows you next
4. You can tap on every product, see its information, and add it to a cart to buy it
AI is eating the world.