exec @ContextVector

Today, we’re excited to introduce Miso One, the most emotive voice model in the world. Miso One is an 8-billion-parameter text-to-speech model for highly expressive speech generation. It emotes like a human and responds faster than a human, with just 110 milliseconds of latency. We’ve open-sourced the model weights, with API access coming soon. Hear how Miso One sounds in the thread below.

Your CEO should be strong. Your CTO should be wise. Your COO should be wicked, cunning, of mysterious origins, fluent in the dark arts, blurry in pictures,

BREAKING: Engineering team ships 50,000 lines of AI-generated code in one sprint. QA team files resignation in the next.

aiCompaniesRightNow

Lmao The glass fiber shortage has even spread to fiber preform All the optical fiber drawn out from a parent rod or fiber preform

Zephyr @zephyr_z9

3 months ago

Btw, the price of single-mode fiber (drones and AI DCs) has increased a lot since December China is the biggest market, and they are seeing a rapid price increase G657.A2 grade bend insensitive fiber (drones & AI DCs) went from 35RMB/km to 70RMB/km in 2 months G657.E grade fiber

5 7 98 64K 39

It's starting. The company with the cheapest coding model at a good enough quality is set to win big, as devs + companies become more price sensitive. Of all the major players, Cursor is in a v good position with their Composer model. Plus eg Factory with smart routing

Ehsan @acadictive

13 hours ago

Dear GitHub Copilot team, I am happy to announce that I successfully burned all of my monthly tokens in under 3 days thanks to your garbage new pricing model. I'd also like to inform you that I won't be renewing my subscription or adding more budget. Best, A former customer.

165 159 2K 314K 100

Dear GitHub Copilot team, I am happy to announce that I successfully burned all of my monthly tokens in under 3 days thanks to your garbage new pricing model. I'd also like to inform you that I won't be renewing my subscription or adding more budget. Best, A former customer.

Gemma 4 12B can now run locally on just 8GB RAM via Dynamic GGUFs. Google's new model, Gemma 4 12B Unified supports image, audio and 256K context. You can run and train the model via Unsloth Studio. GGUF: huggingface.co/unsloth/gemma-… Guide: unsloth.ai/docs/models/ge…

Google Gemma @googlegemma

15 hours ago

Meet Gemma 4 12B! A unified, encoder-free multimodal model designed to bring high-performance intelligence directly to your laptop, and released under an Apache 2.0 license. Bridging the gap between edge efficiency and advanced reasoning. Here is what’s new with Gemma 4 12B: 👇

301 1K 10K 2.1M 4K

Meet Gemma 4 12B! A unified, encoder-free multimodal model designed to bring high-performance intelligence directly to your laptop, and released under an Apache 2.0 license. Bridging the gap between edge efficiency and advanced reasoning. Here is what’s new with Gemma 4 12B: 👇

In early May, the best superforecasters predicted that, by the end of the year, the longest METR 80% task horizons would reach 3-4 hours. In late May, Claude Mythos achieved that number.

Forecasting Research Institute @Research_FRI

2 days ago

We also asked forecasters to predict the longest 80% success time horizon achieved by the end of 2026. All three groups had medians between 3 and 4 hours, up from a baseline of 1.5 hours at the time of survey launch. During our survey period, METR updated its benchmark to

1 3 37 43K 10

Hi. Over the last 24 hours we had three separate small incidents that affected Codex reliability. Those are three too many and we are taking active steps for them to not reproduce. I have reset usage limits for Codex across all paid plans. May the tokens flow again.

Sam doesn’t have any excuses Why aren’t you posting more?

Sam Altman @sama

2 days ago

one of the quotes i find most inspiring on a hard day: "Whatever your hand finds to do, do it with all your might, for in the realm of the dead, where you are going, there is neither working nor planning nor knowledge nor wisdom" Ecclesiastes 9:10

1K 3K 19K 1.7M 6K

Tout est maintenant open-source. Les agents LLM ont atteint un niveau où ils sont capables de prendre un binaire, le décompiler et reconstituer en autonomie le code source du programme. Aucune compétence technique requise. Suffit de leur donner les bons outils (ici, ghidra-mcp)

keyboard_arrow_left Previous keyboard_arrow_right Next

Nicolas @nb4ld

2 days ago

Bienvenue en 2026. Considérez que tout ce qui est dans la nature est maintenant hackable, décompilable, reverse-engineerable en quelques heures. (J'aime)

2 20 318 124K 182

Visual AI is shifting from pixels to code. With Arrow 1.1, every illustration is generated as editable SVG source code, creating outputs that can be refined, adapted, and integrated into real workflows.

a16z @a16z

2 days ago

The most interesting visual AI tools today are generating the underlying source code behind the final output. This change is unlocking editability, iteration, and a feedback loop that pixel-native models can't match. And the market for visual code generation is organizing

61 99 794 199K 865

This is what happens when u have SOTA cache hit rate & attention mechanism Very excited for V4.1

Ara Kharazian @arakharazian

16 hours ago

NEW: DeepSeek, the Chinese AI company, is one of the fastest growing vendors on Ramp. In probably the biggest sign that companies are looking for cheaper alternatives to OpenAI and Anthropic, some are willing to use cheaper, Chinese models, sending U.S. data back and forth from

11 38 273 93K 125

Very impressive results from Min Li and @Haoxiang__Wang: simply swapping Attention for Parallax reaches 2880 steps with the SOAP-H optimizer, beating the latest SOTA record on modded-nanogpt (@kellerjordan0) with no hyperparameter tuning. A few observations: - Parallax is uniformly stronger than Softmax Attention across all records. - Optimizers don't transfer to Parallax with the same magnitude, which confirms the optimizer–architecture interaction from the Parallax paper. - The cleanest modifications often transfer best; records built on heavy tuning transfer less reliably. These are preliminary results, I believe both the Parallax architecture and the optimizer side have room to improve. Code is open-sourced below, give it a try. Code: github.com/Yifei-Zuo/modd… Kernel: github.com/Yifei-Zuo/Para… Paper: arxiv.org/abs/2605.29157

keyboard_arrow_left Previous keyboard_arrow_right Next

Late 2030s, early 2040s tech

Object Zero @Object_Zero_

a day ago

Single Crystal CVD Diamond Have no doubt, you are at the dawn of an industrial revolution. There is a string of breakthroughs happening throughout upstream industries that all compound. Diamond manufacturing is now able to produce CPU size single crystals wafers. Currently

61 244 2K 508K 2K

Single Crystal CVD Diamond Have no doubt, you are at the dawn of an industrial revolution. There is a string of breakthroughs happening throughout upstream industries that all compound. Diamond manufacturing is now able to produce CPU size single crystals wafers. Currently these are marketed as heat spreaders because they have thermal conductivity of 2,200 W/mK which means they move heat incredibly effectively. However, that somewhat misses the wood for the trees… Diamond has physical and electrical properties that exceed traditional silicon, making it uniquely suited for high demand applications. Thermal Conductivity: Heat is the enemy of electronics. Diamond conducts heat better than almost any other known material, about 5 times better than copper and over 10 times better than silicon. A diamond chip can act as its own heat sink. Ultra Wide Bandgap: Diamond can handle massive amounts of voltage and operate at incredibly high temperatures without electrical breakdown. This makes it perfect for high power applications like electric vehicle inverters, power grids, and aerospace technologies. High Frequencies: Electrons move very quickly through diamond, allowing chips to operate at much higher frequencies, which is ideal for advanced telecommunications and radar. Radiation Hardness: Diamond is incredibly resilient to radiation, making diamond based chips ideal for satellites, space exploration, and nuclear facilities. To make a material act as a semiconductor, you have to "dope" it. To do this you inject impurities into the crystal lattice to create a positive (p-type) or negative (n-type) charge. Diamond's atomic structure is so tightly packed that forcing other elements into it is hard. While p-type doping (with boron) has been figured out, reliable n-type doping (with phosphorus) remains a massive hurdle. Theoretical ceilings Band gap Silicon wafer = 1.1 eV Diamond CVD wafer = 5.5eV Clock speed Silicon wafer = 5-6 GHz clock wall Diamond CVD wafer = 1-2 THz clock wall Max Running Temp Silicon wafer = 150°C Diamond CVD wafer = 1,000°C Whilst we etch silicon with photolithography and Extreme UV light, this doesn’t really work with chemically inert diamond. Diamond CVD is currently etched with oxygen plasma etching, but this lacks the precision of EUV. However, we can etch diamond to extreme precision with electron projection lithography. EPL was invented in the 90s by Bell Labs, IBM and Nikkon but abandoned as it was harder than EUV. Electrons repel each other so the beams blurrs too readily. What if we built a femto electron beam? What if we built it to extreme such that it was a ‘single electron’ pulse? What if we build a microscopic "bed of nails" containing millions of nanoscale tungsten or silicon tips (photocathodes). You shine a massive, highly complex femtosecond laser system across the entire array. Every time the laser pulses, millions of tiny tips each fire a single, perfectly straight electron at the exact same time. Turns out, research teams at likes of MIT and Stanford are currently experimenting with exactly this, laser driven nanotip electron emitters. Pair that tool with Diamond CVD substrate tech and we approach the material limits of both semiconductors and nanotechnology. Would require asynchronous logic to escape fatal clock skew and operate at full capability. But I think I will live to see it.

Introducing Wall Attention. Diagonal forget gates enable RoPE-free attention with exceptional length generalization. Wall outperforms the dominant method RoPE and sophisticated data-dependent methods like Forgetting Attention (FoX). We trained models with Wall on 4k sequence length and they generalized without further training to 200k+ tokens. Wall generalizes diagonal forget gates from linear RNNs (KDA, RWKV 7, GLA) to softmax attention through a principled induced action framework. It enables transformers to selectively remember or forget per-channel within the attention head, dramatically boosting expressivity. Wall is production-ready. Wall retains the parallel structure of vanilla attention, is compatible with GQA & MLA, and we open-source reference Triton kernels for training and decoding. Our WallDecode kernel achieves SOTA-level decode throughput. Continual learning over long-context is fundamentally about selective forgetting → and Wall attention is all about selective forgetting.

Tilde @tilderesearch

2 days ago

x.com/i/article/2061…

7 41 356 80K 349

Fyodor Dostoevsky

Electra @Electrarythm

3 days ago

People who think they are extremely self-aware are often doing nothing more than turning themselves into a permanent psychological project. Every reaction is analyzed, every emotion is interpreted, every insecurity is examined, every childhood event is revisited. They call it

140 2K 17K 960K 7K