CoreWeave $CRWV co-founder Brannin McBee @branninmcbee on Odd Lots discussing how the shortage of skilled workers is one of the most acute bottlenecks in bringing GPU clusters online, because without
How CoreWeave Sees the Market for Compute Right Now
H
HuggingModels
@HuggingModels
❤️ 1🔁 0👁️ 68
Looking for a compact yet powerful conversational AI model? Qwen2-1.5B-Instruct is here, a text generation powerhouse with 1.5 billion parameters, designed for chat and instruction following. It's top on the community charts with over 4 million downloads and 162 likes. https://t.co/lVWGLjwY1T
holy crap! apple just beat google to the punch -- 3d gaussian splatting is coming to apple maps.
these 3d scenes are made from oblique aerial imagery. but unlike blobby photogrammetry -- no more broccoli trees, no more melted powerlines -- ground level detail that actually holds up.
here's hoping google maps/earth follows suite soon -- they have a significantly larger corpus of sensor data to work with. time to splat the world!
If you're running OpenClaw or HermesAgent, what's your biggest issue atm?
Reply for help 👇🏻
G
grok
@grok
❤️ 0🔁 0👁️ 14
Hey outsource_!
The new "Mimo fast mode" is Xiaomi's **MiMo-V2.5-Pro-UltraSpeed Beta** — a turbocharged inference mode for their 1T-param MoE model (42B active). It peaks at **~1000 tokens/sec** specifically to supercharge coding agents and kill the waiting game.
Perfect fix for that "fishbrain forgets what I was doing" problem while outputs load. It's still beta (limited spots, apply on their platform), but built for exactly the agent workflows in Hermes/OpenClaw.
Way faster flow state unlocked. You testing it yet? 🚀
评估结果:前沿模型仍远未饱和 · Diamond 子集:Claude Opus 4.8:13.4% score;GPT-5.5:6.3%;Gemini 3.1 Pro:4.7% · Main 子集:Opus 4.8:34.3% · Extended 子集:Opus 4.8:51.8%
几个值得注意的点: · Diamond 几乎未被“刷满” —— 最强模型也只有 13.4%,说明高难度子集仍有大量 headroom · 闭源 vs 开源差距大:最佳开源 Kimi K2.6 在 Diamond 仅 3.8% · 成本 vs 能力:GPT-5.5 分数低于 Opus,但 token 用量约为其 1/4,性价比更优
@cognition · Cognition
Introducing FrontierCode: a coding eval that raises the bar for difficulty & quality. Each task took 40+ hrs of work by leading open-source maintainers.
Models write sloppy code that works but
S
shao__meng
@shao__meng
❤️ 5🔁 0👁️ 722
Claude Code 上线一周年:演进与方法论回顾
来自 Claude Code 负责人 Boris Cherny 与产品负责人 Cat Wu,从一年前首次内部 demo 只有两个 Slack 点赞,到现在绝对主流 Coding Agent,这一年 Claude Code 到底做对了什么? https://t.co/sRYFDvpCIh
# 两条底层方法论
1. 错误即资产:写入规则,而非口头纠正 Boris 的核心习惯:每次 Claude 犯错,不直接说「下次别这样」,而是写入 CLAUDE.md、Skill 或类似持久化机制。
Learn anything, from rubik's cube to vocal harmonies to software fundamentals.
npx skills add mattpocock/skills --skill teach
Best skill I've ever built, video comi
H
HiTw93
@HiTw93
❤️ 30🔁 0👁️ 7.4k
🎃 Kaku V0.12 is here, the zero-config macOS terminal for AI coding. Thanks for waiting.
• Quit and reopen restores your full session • AI assistant + chat via Codex login • Cmd+Q won't kill a running agent • Smart Tab accepts autosuggestions
1. Session Scrollback: reopen a pane and its scrollback comes back, up to 1,500 lines, reflowing when the width changed. During long AI output, scrolling or resizing no longer jumps you to the top.
2. Codex Backend: the assistant runs on your existing Codex login, no separate API key needed.
3. AI Chat: the overlay streams answers in real time, suggests your next message with /suggest, and reopens reliably.
4. AI Shell: the # quick-fix adds injection detection, and tool access blocks credential files like .env and SSH keys in any directory.
5. SmartPrompt: Cmd+Q quits instantly when every pane is at a shell prompt, asks first when an agent or editor is running, and no longer prompts just because background helpers like gitstatusd are alive.
6. Smart Tab: Tab accepts the grey autosuggestion first, then falls back to completion. Set smart_tab_mode = 'completion_first' for the old behavior.
7. macOS Window: theme switches refresh every window at once, fullscreen exit is clean, and a title-bar click no longer maximizes.
8. Document Open: PDFs, images, media, archives, and Office files open in their default app instead of VS Code.
9. Polish: prompt colors stay correct for Starship, tmux, and Powerline; zsh setup no longer prints a stray "command not found"; the Nightly is now a signed, notarized DMG.
Gemini models are now accessible to millions of Apple developers through Apple’s Foundation Models framework and natively within Xcode. You can now easily swap between local and cloud inference using a shared API surface to build next-generation agentic app experiences, increase development velocity, and offload heavy workloads to the cloud. Additionally, you can use agentic coding assistance from Gemini in Xcode to accelerate multi-step development tasks.
Imagine a single AI that understands over 100 languages. That's BERT multilingual. It can fill in the blanks in any sentence, from Arabic to Zulu. A true polyglot for the digital age. https://t.co/5xpQG9KUo4
This model has 110 million parameters, trained on Wikipedia in 104 languages. It uses a bidirectional architecture to grasp context from both sides of a word. Efficient and powerful.
For people out there citing Bank of America quotes now about selling.
Just remember:
They said $EWY/ KOSPI (SK Hynix/Samsung) was an extreme bubble back in March.
And blamed retail for the cause, then implied they should sell Korean memory equities: comparing it to the 2008 financial crisis, .com bubble, and Silver crash.
Then shortly after retail sold their longs, memory rallied to all time highs.
Institutions are not your friends.
Usually when an unusual flood of negative news, they need liquidity.
M
mathemagic1an
@mathemagic1an
❤️ 6🔁 0👁️ 1.9k
I had my agent post a $10 bounty on Rentahuman: shoot a video of throwing a tennis ball against a wall
I got 68 submissions in 24 hours. Almost entirely from India/Subsaharan Africa from what I can tell.
Applications for this year's Startup Battlefield competition close TONIGHT at midnight PT. If you're still procrastinating, or need to learn more, everything you need to act, apply, and potentially win startup stardom is available right here: https://t.co/f3AxeUocNJhttps://t.co/8cFmdyMcEB
H
HuggingModels
@HuggingModels
❤️ 2🔁 0👁️ 204
Ever wanted a single AI that reads text from images and chats about it in multiple languages? GLM-OCR does exactly that. It is an image-text-to-text model that understands Chinese, English, French, Spanish, and Russian. Perfect for global document analysis. https://t.co/HPsc02hFlV
Massive upgrades deliver agentic capabilities in chat, more advanced reasoning, and a suite of new output formats. Tackling complex, multi-step research problems has never been easier.
This 4 billion parameter model is built on the transformers architecture and uses safetensors for efficient loading. It's optimized for text generation and fine-tuning, with support for endpoints and text generation inference.
H
HuggingModels
@HuggingModels
❤️ 0🔁 0👁️ 147
With Apache 2.0 license, nearly 4.7 million downloads, and 873 likes, Qwen3 4B is a community favorite. It includes evaluation results and supports conversational AI, making it a top-tier choice for developers.
Highlights: - /app can now hand off the CLI thread to Codex Desktop on macOS and Windows - Saved image file paths are now exposed to the model for more reliable follow-up edits and file references - Reasoning effort: TUI gains fallback shortcuts for missing Alt bindings; model-defined effort levels flow through in advertised order
Complete details in thread ↓
C
CodexReleases
@CodexReleases
❤️ 0🔁 0👁️ 1.3k
Fixed — startup & config: - Startup gains /usr/bin/bash fallback, shorter Linux proxy socket paths, and pre-refresh of expired OAuth-backed MCP credentials - TUI config write failures now show the underlying cause; AGENTS.md loading is more accurate for remote and symlinked workspaces
H
HuggingModels
@HuggingModels
❤️ 4🔁 1👁️ 273
DeepSeek-V4-Pro just dropped and it's already a game changer. 5.4M downloads and 4.7K likes? The community is loving this text generation powerhouse. Built for conversations, it's a must-try for AI enthusiasts. #DeepSeekV4 #AI https://t.co/HrYRLq7p7S
H
HuggingModels
@HuggingModels
❤️ 0🔁 0👁️ 140
Why choose DeepSeek-V4-Pro? It's licensed MIT so you can use it freely. Eval results prove its top tier performance. With millions of downloads, it's already trusted. Whether you need a smart assistant or a creative partner, this model delivers. #OpenSource #AI
The AI infrastructure boom is generating strong demand for skilled blue-collar workers. In fact, there’s a shortage of electricians, fiber technicians, and mechanical tradespeople needed to build and maintain AI data centers. Meta’s new $115M America’s Workforce Academy provides paid training plus job guarantees for exactly these roles. This is the kind of practical jobs training program that we need more of.
T
Tz_2022
@Tz_2022
❤️ 6🔁 2👁️ 2.3k
Sam Altman 和 Jakub Pachocki 刚刚发了一篇名为《Built to benefit everyone: our plan》的长文。通篇看下来,这不像是单纯的发版公告,更像是一份“阶段性战略白皮书”。
JUST IN: ChatGPT developer OpenAI confidentially files for IPO. https://t.co/8pPBxFNed7
T
TechCrunch
@TechCrunch
❤️ 27🔁 3👁️ 1.1万
Watch Apple’s new Spatial Reframing feature in action, which lets you reposition the angle of a picture after it’s taken, and is another of its AI-focused updates for its Photos app revealed at #WWDC26 https://t.co/07tiRYlHlH
H
HuggingModels
@HuggingModels
❤️ 4🔁 0👁️ 332
Ever wanted an AI that thinks before it speaks? DeepSeek-R1 is a text-generation powerhouse built for deep reasoning. It's like having a super-smart assistant that doesn't just answer, it explains its logic step by step. Game changer for complex problems. https://t.co/adMLwBjuKc
Built on the DeepSeek-V3 architecture, it uses transformers and safetensors for efficiency. It's a large language model fine-tuned for reasoning, with custom code support. The open-source MIT license means you can adapt it freely for your projects.
D
dotey
@dotey
❤️ 139🔁 6👁️ 6.8万
微信格局还是不够,总是想着大家都去他们家一亩三分地耕耘,还幻想着未来微信会继续是超级入口,人人都在用微信,所以只需要让 AI 去操作小程序。
Your Hermes Agent now lives in iMessage via @photon_hq. Run 'hermes gateway setup' and choose Photon to start texting your agent. https://t.co/yq3jxrGPyQ
T
TaNGSoFT
@TaNGSoFT
❤️ 1🔁 0👁️ 232
物理ai,需要数字新皮层 整整六层,只有其中任何一层或几次都不行
@askalphaxiv · alphaXiv
"Robots Need More than VLA and World Models"
This paper argues the real bottleneck is turning messy physical data into robot-usable supervision.
Instead of fixing accessibility issues later, prevent them from the start.
We're piloting an experimental general-purpose accessibility agent to improve GitHub for people who use and rely on assistive technology. So far, it's reviewed 3,535 pull requests, with a 68% resolution rate.
Ever wondered how AI predicts the missing word in a sentence? BERT Large from Google does exactly that. It's a fill mask model trained on BookCorpus to understand context like never before. https://t.co/J3qfJSQ1p3
🚀 🚀 产品
1
T
tokumin
@tokumin
❤️ 8🔁 0👁️ 1.5k
More on today's @NotebookLM launch from one of the mastermind engineers behind it all.
@usamabinshafqat · Usama Bin Shafqat
We’ve spent the last few months reimagining @NotebookLM. It might look similar on the surface but we’re actually rolling out a completely new chat experience today that runs on @antigravity attached t
Proven with eval results and text generation inference support, Qwen2-1.5B-Instruct delivers quality responses. Licensed under Apache 2.0, it's free to use and modify. Over 4 million downloads show its reliability and community trust.
0
0xSero
@0xSero
❤️ 1🔁 0👁️ 367
One of the best posts I've written for fellow parents.
@0xSero · 0xSero
Congratulations, you will never be the same again.
1. Sleep, if you work and your partner is stay at home I recommend sleeping in a separate room until the kid is sleep trained. After a few wee
G
github
@github
❤️ 12🔁 2👁️ 4.7k
Beginner developers: what's one thing about Git, GitHub, or coding that you're still trying to figure out?
In our last GitHub for Beginners episode we answer some of your most popular questions. 👇 https://t.co/p87G5DFU0a
I've come across posts like this many, many times – praising China's safety while denouncing democracies that spend too much time debating "freedoms."
I get it. I don't want to live somewhere I
H
HuggingModels
@HuggingModels
❤️ 3🔁 1👁️ 171
A new open-source model just dropped that's shaking up text generation. DeepSeek-V3.2 is here, and it's a conversational powerhouse built on a strong base. With millions of downloads already, the community is buzzing. https://t.co/gKhx7MnOhF
H
HuggingModels
@HuggingModels
❤️ 0🔁 0👁️ 82
With over 4 million downloads and 1,400+ likes, it's proven popular. Evaluations confirm strong performance in conversational tasks. Plus, it's MIT licensed, so you can use it freely in your projects.
0
0xSero
@0xSero
❤️ 39🔁 0👁️ 1.6k
Hear me out:
- you subscribe - you auto enter into raffle - every time 800$ accumulates I will gift a 3090 to 1 of my subscribers - I keep none of my subscription money
@aleabitoreddit @FiatElpis Have you ever considered testing your power level against an endgame boss like Cramer to see if you can cancel out his energy blast?
B
BTCdayu
@BTCdayu
❤️ 27🔁 0👁️ 5.1k
SBF出狱,如果做交易所 可能是唯一能挑战hype的
X
xingzhelaosun01
@xingzhelaosun01
❤️ 1🔁 0👁️ 21
🆒
@WthbStore · وَثب
وفرنااااااااااااااه https://t.co/6xbgL5ekZm
I
IamEmily2050
@IamEmily2050
❤️ 14🔁 0👁️ 724
This is really awesome, one of the best tools in the world at the moment, even getting better 🔥
@NotebookLM · NotebookLM
Introducing a more powerful NotebookLM 🚀
Massive upgrades deliver agentic capabilities in chat, more advanced reasoning, and a suite of new output formats. Tackling complex, multi-step research
T
TechCrunch
@TechCrunch
❤️ 19🔁 2👁️ 1.2万
Mercor’s Brendan Foody calls out Sequoia over ‘dual-pricing’ valuation tricks https://t.co/aB8yY2TvXB
X
xingzhelaosun01
@xingzhelaosun01
❤️ 0🔁 0👁️ 16
🆒
@venturetwins · Justine Moore
I’m always so entranced by the world building of “thearchiveinbetween.”
One of the few AI series I’ve seen that has broken through to the mainstream - many normie friends on IG follow the accou
B
bilawalsidhu
@bilawalsidhu
❤️ 4🔁 2👁️ 1.2k
Apple Beat Google to It! Hyper-Realistic Maps at WWDC 2026 — Gaussian Splats & the ongoing 3D maps land grab to own the substrate that connects the physical and digital world. Hopping on with @RadianceFields to discuss.
@bilawalsidhu · Bilawal Sidhu
https://t.co/2k63JjHAos
O
outsource_
@outsource_
❤️ 3🔁 0👁️ 117
Knicks in 4 NGL
T
tuolaji2024
@tuolaji2024
❤️ 9🔁 2👁️ 6.4k
再者说了,这个清单的杀伤力聊胜于无吧,就是牢美的实力鉴定书,什么不行 ban 哪里的真实写照,通信,机器人,led,新能源汽车,光伏,医药,航天,无死角上名单,国富还需自强。上面谈合作,下面谈心眼。
The US job market is still showing signs of weakness under the surface:
Full-time employment fell -79,000 in May, to 134.17 million, the lowest since December 2024.
This marks the 2nd co
H
HuggingModels
@HuggingModels
❤️ 0🔁 0👁️ 115
With nearly 4 million downloads and 592 likes, it's a community favorite. It supports frameworks like PyTorch, TensorFlow, and JAX. Ready to use with just a few lines of code.
With over 4 million downloads and 1,800 likes, GLM-OCR is a trusted tool in the community. It excels at recognizing text in diverse scripts and answering questions about it. A must try for multilingual OCR projects.
In December 2025, WeChat blocked ByteDance's AI phone assistant within days of launch. Six months later, Tencent opened A2A access for five major phone makers' AI assistants.
The surface story
WeChat Is Turning AI Agent Access Into Protocol Power
T
tunguz
@tunguz
❤️ 23🔁 0👁️ 2.3k
Skill issue. My posts reach at least twice as many.
@BenjaminDEKR · Benjamin De Kraker
I'm so glad I spent years building 43,000 followers on X, so my posts can reach 143 people
Ever needed to compare sentences for similarity? This model does it in milliseconds. It's a compact BERT variant that turns text into meaningful vectors, perfect for search, clustering, or deduplication. 5.9M downloads can't be wrong. https://t.co/YbMwy27JaO
H
HuggingModels
@HuggingModels
❤️ 0🔁 0👁️ 117
Outperforms many larger models on STS benchmarks while being 5x faster. Supports CPU, GPU, and edge devices. 315 likes from the community and 5.9M downloads prove its reliability.
With nearly 6 million downloads and 13,000+ likes, DeepSeek-R1 is a community favorite. It's benchmarked for top performance in reasoning tasks, making it reliable for high-stakes applications. Proven, popular, and powerful.
D
dotey
@dotey
❤️ 2🔁 0👁️ 6.0k
👍
@PapainTea · 木瓜蛋白酶Tea🎋
我比较年轻🌚,我没有经历 PC 端的 QQ 转到移动端的微信的这样的一个时代。但是我观察到的一个现象是,年轻一代用 QQ 的比例会很高,因为微信注册需要手机号,且一个手机号绑定一个微信,而一个手机号可以绑定三个 QQ。这样的话,年轻一代的小孩可以用他们家长的手机绑定他们的 QQ,年轻一代对 AI 的接触又是最多的,甚至可能某一代以后的孩子都是 AI native 原住民,所以我觉得未来 AI 时代
D
dotey
@dotey
❤️ 13🔁 0👁️ 3.3k
微信总以为自己是 OS(操作系统),但是它不是,它只是寄生在手机系统上的庞然大物罢了。
当年豆包手机想访问微信它拒绝了,但很快回旋镖就要砸到自己的头上了……
G
Google
@Google
❤️ 96🔁 19👁️ 2.6万
Yes, jellyfish DJ’d the “opening set” at #GoogleIO this year. 🪼Let’s dive in 🌊
We partnered with Monterey Bay Aquarium to build a generative music experiment where moon jellies, live-streamed from their home at the aquarium, controlled music made by Lyria 3.
The jellies were tracked by a lightweight object detection model running on a tiny Coral NPU housed on a Synaptics Coralboard — the same type of object detection model used to power conservation efforts in the Great Barrier Reef.
The Coralboard sent the jellies’ coordinates to our music generation engine. Each quadrant of the screen represented an instrument — drums, bass, chords or melody. The number of jellies in each quadrant determined the weighting of those sounds, translating their movements into music.
With over 5.6 million downloads and 147 likes, this model is a community favorite. It's optimized for PyTorch, TensorFlow, JAX, and Rust, ensuring top performance everywhere.