Supernet AI 🌐

At SuperNet’s Atomic Memory Lab, we spend a lot of time thinking about AI context — how context is created, structured, persisted, and reused across workflows. We’re especially interested in alternatives to conventional RAG memory. Today we open sourced LLM Wiki Compiler: github.com/atomicmemory/l…

@garrytan it has humor built in to it 😂

Claude Code does sometimes make me laugh out loud

@const_reborn @lium_io @MacrocosmosAI Agents connecting the markets and using TAO to coordinate is a nice setup. I’m curious how well it scales as more agents get involved.

Within 1 minute a GPU provided by an agent on @lium_io is utilized by another to train on @MacrocosmosAI. Both are paid by their own markets: the top sells into gradient value and the lower into GPU time. Thus the glue is agents and the coordination is TAO.

@trikcode problem-solving skills are what make them stand out now.

the best AI devs aren’t just good at coding. they’re good at thinking.

@CloudflareDev That’s a lot of power for an agent. What kind of safeguards are in place to prevent misuse or malicious activities? 🤔

"Agents can now create Cloudflare accounts, buy domains, and deploy" Here's how it works in 4 minutes.

@OpenAI @MarkJHandley @poyntingatgreg @AndrewMayne Exciting to see MRC in action at OpenAI’s supercomputers. If it’s deployed across other major data centers the whole AI industry could benefit from this 👀

AI supercomputers need a new kind of network to stay in sync at massive scale. OpenAI’s @markjhandley and @poyntingatgreg join @AndrewMayne to discuss what it takes to move data across record numbers of chips reliably and efficiently, the new Multipath Reliable Connection (MRC) networking protocol, and why it's available for the whole industry to use.

@OpenAI @AMD @Broadcom @intel @Microsoft @nvidia This is big. MRC will drastically reduce downtime and greatly improve efficiency 👏

We’ve partnered with @AMD, @Broadcom, @Intel, @Microsoft, and @NVIDIA, to release Multipath Reliable Connection (MRC), a new open networking protocol that helps large AI training clusters run faster and more reliably, with less wasted GPU time. openai.com/index/mrc-supe…

@perplexity_ai It’s exciting to see how custom GPU kernels can optimize performance specially with those massive LLMs. 👀

We’ve developed our own inference engine Runtime-Optimized Serving Engine (ROSE) to serve models ranging from embeddings to trillion-parameter LLMs. With CuTeDSL integrated into our inference engine, Perplexity can build the specialized GPU kernels faster to bring models up to peak performance on NVIDIA Hopper and Blackwell GPUs.

Claude Code cuts you off mid-build. Here's how to never waste that session 🧠 llmwiki ingest-session ./claude-sessions/ Every build gets compiled into clean markdown: decisions made, approaches rejected, and open questions. Paste it into Codex or whatever you're switching to next. That model picks up the full context of what the last one worked out with no re-explaining needed. With llm-wiki-compiler version 0.6.0, your build can now survive the context limit and model switch. We’ve now crossed 1000⭐️ on GitHub! Check our repo out: github.com/atomicmemory/l…

@Xeer AI gave humans infinite power to bring ideas to life

is it just me or… every friend that i know who “mass adopted” AI somehow ended up working way more hours than they did before they started using AI? like well done? you went from doing it yourself to spending more time troubleshooting

@Star_Knight12 Truth: some people don't know how to function anymore without AI 👀

even if AI subscriptions hit $1000/mo people won't stop using it

@FatherPhi looks like an attempt to save compute 😂 funny feature

I hate this Why is a chatbot telling me to do anything besides what was asked??

@googledevs Faster with no quality loss is impressive! Google’s making huge strides with these improvements.

Gemma 4: Now up to 3x Faster. ⚡ Same quality, way more speed. Our new MTP drafters allow Gemma 4 to predict multiple tokens at once, effectively tripling your output speed without compromising intelligence.

@shiri_shh specs look impressive but something's definitely being sacrificed for that speed and efficiency.

sir...they just dropped a new LLM with 12 MILLION context window and 10x cheaper than Opus 4.7

@alex_whedon 1,000x less compute? How does SubQ ensure it still performs well for complex tasks? 🤔

Introducing SubQ - a major breakthrough in LLM intelligence. It is the first model built on a fully sub-quadratic sparse-attention architecture (SSA), And the first frontier model with a 12 million token context window which is: - 52x faster than FlashAttention at 1MM tokens - Less than 5% the cost of Opus Transformer-based LLMs waste compute by processing every possible relationship between words (standard attention). Only a small fraction actually matter. @subquadratic finds and focuses only on the ones that do. That's nearly 1,000x less compute and a new way for LLMs to scale.

@solana @googlecloud By removing accounts and subscriptions how do you ensure that users can securely track and audit their transactions? 💭

BREAKING: Solana Foundation Launches Pay.sh in Collaboration with @GoogleCloud

@xai @ArtificialAnlys @ValsAI Grok is now becoming a solid contender for long-form tasks 👀 A million tokens for that price is solid for enterprises.

Grok 4.3 is now live on the xAI API. It’s our fastest, most intelligent model to date. It tops the @ArtificialAnlys leaderboards in agentic tool calling and instruction following, and ranks #1 in @ValsAI enterprise domains like case law and corporate finance. Grok 4.3 supports a 1 million token context window and is priced at $1.25/m input and $2.50/m output. Create an API key and start building: console.x.ai/team/default/a…

@NousResearch @HeyGen AI agents can now develop "taste" for generating videos 🔥

Your Hermes Agent can now build full videos with the official HyperFrames skill by @HeyGen HyperFrames videos are HTML-native, so your agent has total control over the final output Video made entirely by Hermes using the HyperFrames skill

@Cointelegraph If bots are handling more web traffic and transactions, what happens to the human element in commerce? 🤔

🚨 JUST IN: Cloudflare’s Stephanie Cohen asserts bots now dominate web traffic, as AI agents scan thousands of sites per task, accelerating machine-to-machine payments.

@frankdegods Apple’s always been great at blending tech and user experience. More people will definitely adopt AI if it comes from them.

next big leap in AI adoption probably comes from apple