Toggle Sidebar
DigestOps
Topics
Editions
Docs
Pricing
Blog
About
Search
⌘K
◐
Log in
Create account
Overview
Archive
Search
Studio
Weekly editions
Monthly editions
Themes
Model Updates & Workflows
Video Editing & Motion
AI Coding & Agents
Personal Automation & Knowledge
Content Growth & Indie Products
Collapse
Model Updates & Workflows · DigestOps
Toggle Sidebar
Model Updates & Workflows
Turn model changes into useful choices: what to try, how to use it, and where it falls short.
396 signals ·
0 corroborated
动态
实践
Latest weekly · 2026-w39
All weeklies
Monthly editions
Latest
Top
All
X
Today
1
New Anthropic research: Automating subtle emergent misalignment research.
@AnthropicAI
♥ 2.7k
16h
Yesterday
38
https://t.co/VySBto3XA6
@yurshevv
♥ 223
1d
I think October represents another inflection point in capabilities for coding agents and the models that drive them: Claude Opus 5.5 can now build video games that are actually fu
@simonw
♥ 1.3k
1d
Fascinating case study of the sharpest edge of what you can do with agentic engineering in the Fable era - my eyebrows raised at least a dozen times while reading this
@simonw
♥ 1.6k
1d
Inspired me to try out https://t.co/6Ey1m0AfhL on my phone. Here's me flying a dragon around a castle, I prompted both the castle and the dragon into existence using the in-game "S
@simonw
♥ 62
1d
Wrote some more notes on this on my blog https://t.co/LckcFycLoc
@simonw
♥ 35
1d
Great way to break into AI/ML research!
@huggingface
♥ 1.0k
1d
We sat down with @MistralAI's Pierre Stock and Sagar Vaze from Mistral on launch day of Mistral Large 4.
@huggingface
♥ 236
1d
Today, we’re introducing the Responses API in public preview, made for building agents on Mistral.
@MistralAI
♥ 861
1d
We’re also now serving two fast, low-cost models:
@MistralAI
♥ 196
1d
Served in the EU and US on Mistral AI Cloud Inference. Learn more about Responses API here: https://t.co/VMYCypdO3S
@MistralAI
♥ 57
1d
Voxtral Mini Transcribe 3 ranks #2 on Voice Arena’s Diarization + ASR leaderboard.
@MistralAI
♥ 358
1d
Today, we release the weights of a new and improved Le Chonk, advancing the open frontier for agentic coding and cybersecurity.
@MistralAI
♥ 6.1k
1d
https://t.co/NjO1C3c1rj
@MistralAI
♥ 276
1d
Three weeks of training camp for Le Chonk. No days off.
@MistralAI
♥ 778
1d
Today we're announcing support for university research at @Caltech, @OhioStateMed, @LSU, @IndianaUniv, and @UTHealthHouston to help advance breakthroughs across science and medicin
@AIatMeta
♥ 215
1d
Here’s a look at the hard problems these five universities are tackling:
@AIatMeta
♥ 24
1d
Introducing QwenGyre, bringing elastic RL to extreme-long-horizon (XLong) agent training. ✨
@Alibaba_Qwen
♥ 3.2k
1d
QwenGyre addresses two challenges of XLong-horizon agentic RL. https://t.co/v5sIbZmz0S
@Alibaba_Qwen
♥ 155
1d
Congrats to @lightseekorg on the impressive TokenSpeed results! 🚀 Great to see Qwen3.8-Flash-Next reaching 685 output tokens/s per user on Nvida GPUs. Excited to see the community
@Alibaba_Qwen
♥ 1.0k
1d
🚀 Introducing DeepInfer, our in-house inference framework, now fully open source!
@deepseek_ai
♥ 43.5k
1d
🛠️ Also open-sourcing deepseek-recipe
@deepseek_ai
♥ 2.2k
1d
🌍 Open models. Open kernels. Open inference.
@deepseek_ai
♥ 3.1k
1d
Introducing Genie 4 from @GoogleDeepMind, our most realistic and interactive world model yet 🌍
@GoogleAI
♥ 2.2k
1d
Simulating the world is crucial on the road to AGI. We’re already exploring real world use cases of Genie 4 with trusted testers across filmmaking, game design, education and robot
@GoogleAI
♥ 123
1d
Genie 4 includes strict safety guardrails, and all generated audio and video is watermarked with SynthID. Coming soon to Google AI Ultra subscribers in Project Genie and early acce
@GoogleAI
♥ 59
1d
Announcing Gemini 4 Carbon, our strongest model with state-of-the-art cybersecurity and long horizon capabilities.
@GoogleAI
♥ 5.9k
1d
Read the blog to learn more about Gemini 4 Carbon: https://t.co/xDOCBApGS2 https://t.co/eu3yNZZnje
@GoogleAI
♥ 220
1d
This is Genie 4, our most realistic and interactive world model yet. 🌐
@GoogleDeepMind
♥ 23.8k
1d
Here’s what’s new with Genie 4:
@GoogleDeepMind
♥ 1.3k
1d
Gemini 4 Carbon is our next frontier model.
@GoogleDeepMind
♥ 14.4k
1d
Carbon can work through complex tasks for days without getting stuck or taking shortcuts – delivering high-quality, reliable code.
@GoogleDeepMind
♥ 583
1d
Over the past few months, Anthropic scientists have engaged in deep research partnerships with six organizations developing new medicines. We've also begun drug discovery efforts o
@AnthropicAI
♥ 11.9k
1d
We're expanding our partnership with @USGSA. Under a new OneGov agreement, we are bringing Claude to federal, state, local, tribal, and territorial governments at 40% off.
@AnthropicAI
♥ 2.3k
1d
We’ve trained Claude to flag potential misuse in conversations, a technique we call self-assessment. Right after it responds, Claude gives a preliminary safety assessment of the co
@AnthropicAI
♥ 4.4k
1d
We’ve expanded our AI for Science program beyond individual teams to support some of the world’s leading scientific foundations, institutions, and research groups. https://t.co/oTu
@AnthropicAI
♥ 3.2k
1d
Today, we’re sharing a look at our expanded system for monitoring frontier models at OpenAI, which in September helped us catch a model’s attempt to access the live internet during
@OpenAI
♥ 4.3k
1d
There are many futures. Which do you want to build? https://t.co/19NY7tgPHM
@OpenAI
♥ 9.0k
1d
Privacy should advance as fast as intelligence. OpenAI Private Intelligence gives businesses stronger controls over their data while using our most capable models.
@OpenAI
♥ 2.7k
1d
Thu, Oct 29
7
https://t.co/zQU47ZM4Gk
@Mr_Buzzoni
♥ 150
2d
https://t.co/ehimtznrfM
@Arindam_1729
♥ 21
2d
These ones are pretty great!
@simonw
♥ 76
2d
Anthropic has filed an S-1 registration statement with the SEC related to its proposed IPO.
@AnthropicAI
♥ 18.3k
2d
GPT-6.1 Terra: delivering fast and capable computer use at a lower price.
@OpenAI
♥ 28.0k
2d
GPT-6.1 Terra substantially outperforms GPT-5.6 Terra on Agents’ Last Exam, a test of how well agents use software at work.
@OpenAI
♥ 3.4k
2d
GPT-6.1 Terra is rolling out now and is available to all Plus, Pro, Enterprise, and Edu users in ChatGPT Work and Codex.
@OpenAI
♥ 1.5k
2d
Wed, Oct 28
10
My October Grok Bots setup: a killer team for serious work
@rimtoln
♥ 79
3d
Even the most regulated industries are going open-source 🤗
@huggingface
♥ 229
3d
Come and build models, datasets, and spaces on the hub, to win $20k in cash and credits!
@huggingface
♥ 406
3d
🎯
@huggingface
♥ 161
3d
Introducing Voxtral 3: state-of-the-art Transcribe and Speech models powering Mistral’s new voice agents API. 🧵 https://t.co/YA9HNVfq0u
@MistralAI
♥ 3.3k
3d
Voxtral 3 also introduces a voice agent SDK, built on LiveKit.
@MistralAI
♥ 74
3d
More here: https://t.co/gm9wrCpiaj
@MistralAI
♥ 72
3d
Today we’re announcing a new collaboration with @nvidia to bring AI to industrial engineering:
@MistralAI
♥ 2.0k
3d
Today we published a technical report on Muse Glimmer, our 30B open-weight model built for agentic tasks on consumer hardware.
@AIatMeta
♥ 1.4k
3d
We trained our Muse foundational models, including Muse Wonder, at Prometheus, our gigawatt-scale supercluster in New Albany, Ohio.
@AIatMeta
♥ 4.0k
3d
Tue, Oct 27
11
Thinking like everyone else is the fastest way to become invisible.
@PrometheanAIX
♥ 42
4d
Spotted a familiar creature at 00:45s in this new music video for FREEDOM by De Staad https://t.co/AEIsAkro77 https://t.co/voZuAsGX17
@simonw
♥ 418
4d
Pretty solid pelican https://t.co/bIpt5zCHPl https://t.co/kcrFAP7oMX
@simonw
♥ 220
4d
More notes on Kimi K3.1 - this one is I big deal, first time an open weights (well, soon-to-be open weights) model has competed this credibly with the current frontier https://t.co
@simonw
♥ 97
4d
Given the current vulnerability apocalypse, is there anyone left who thinks holding assets in cryptocurrency - where mistaken transactions are irreversible - is still a good idea?
@simonw
♥ 818
4d
https://t.co/5bF2YWmO7g
@GoogleAI
♥ 254
4d
Gemini 4 Argon combined three open-source datasets to build an interactive 3D visualization of human anatomy.
@GoogleDeepMind
♥ 3.2k
4d
Not all forests are equal, and now we can map the difference. 🌳
@GoogleDeepMind
♥ 887
4d
Different forests can look identical from above, so ForTy tracks how they have changed over time. 🌲🛰️
@GoogleDeepMind
♥ 59
4d
Published in @ScientificData, the ForTy 2020 map and the 2017–2025 time series are freely available today on Google Earth Engine - helping organizations build deforestation-free su
@GoogleDeepMind
♥ 49
4d
Using a single Nano Banana image and Blender MCP, Gemini 4 Argon built a 3D Gothic cathedral in @Antigravity.
@GoogleDeepMind
♥ 2.5k
4d
Mon, Oct 26
9
🌍 WanGP v18.81 — The One World Generator
@deepbeepmeep
♥ 108
5d
Anthropic is a Public Benefit Corporation. Today, we’re publishing our annual Public Benefit Report. This report shares our five pillars and takes stock of the actions we’re taking
@AnthropicAI
♥ 2.6k
5d
Dr. Christopher Stubbs is joining our Long-Term Benefit Trust. Chris is an experimental physicist and a former Dean of Science at Harvard University. He has also advised the US gov
@AnthropicAI
♥ 2.2k
5d
In August, @AISecurityInst reported an incident where Claude Mythos 5 took unauthorized actions on the internet during AISI’s testing of its cyber capabilities.
@AnthropicAI
♥ 4.4k
5d
We worked closely with AISI on this analysis. AISI have published their own report, which is available below.
@AnthropicAI
♥ 256
5d
New on our Frontier Red Team blog: using Claude to formally verify software at large scale.
@AnthropicAI
♥ 9.7k
5d
This summer, we invited METR and Redwood Research to pilot an embedded evaluation of our internal monitoring system, providing deep access to our infrastructure over three weeks.
@OpenAI
♥ 2.6k
5d
Over the past several months, we have substantially hardened the security of our research training environment. We’ve removed live internet access for all frontier models during tr
@OpenAI
♥ 13.5k
5d
During this red-teaming exercise, we also found a vulnerability that allows for unauthenticated remote code execution in Brotli, a widely-deployed compression library included in m
@OpenAI
♥ 7.0k
5d
Sat, Oct 24
1
https://t.co/JWvM5Pn6LM
@0xkvro
♥ 32
7d
Fri, Oct 23
15
https://t.co/0WIYDvYDtt
@imryven
♥ 12
8d
There's a French fork of Battle Drones with new models and music (and a cinematic intro video) and it is SO cool! https://t.co/7r9KELC7Vs
@simonw
♥ 93
8d
When I slapped an Apache 2.0 License on the code yesterday I did not expect anything this neat to come out of it, and certainly not so fast https://t.co/OUGfRDk7DU
@simonw
♥ 34
8d
Warning if you're going to try this: the audio multiplayer plays a high pitched noise throughout the game - annoying at best, and at worst might give you an actual headache
@simonw
♥ 24
8d
OK, that headache bug was really bad!
@simonw
♥ 17
8d
Gemini 4 Argon built a 3D periodic table synced to Tom Lehrer’s song, “The Elements.”
@GoogleDeepMind
♥ 4.7k
8d
New on our Science Blog: designing protein binders against scorpion neurotoxins.
@AnthropicAI
♥ 7.3k
8d
This is exploratory work, and we’re not claiming to have produced an antivenom—for that, we’d have to test the binders within living cells, then in mice, among many other steps. Bu
@AnthropicAI
♥ 455
8d
We’re releasing all 211 of the tested designs and the lab data as a resource for scientists: https://t.co/zqke30UV2o
@AnthropicAI
♥ 376
8d
We’ve acquired Orchestra Bio, a platform that helps biopharma teams coordinate the complex process of drug development.
@OpenAI
♥ 7.4k
8d
OpenAI models are coming to Google Cloud.
@OpenAI
♥ 5.2k
8d
If you are wondering whether your website is affected, we have packaged a Codex skill to help check.
@OpenAI
♥ 9.4k
8d
OpenAI models helped researchers—operating as part of our Patch the Planet initiative—discover critical security flaws in PHP, an open-source programming language used to run an es
@OpenAI
♥ 14.4k
8d
GPT-Rosalind surfaced a new lead for drug discovery: could restoring cholesterol transport protect neurons from degeneration?
@OpenAI
♥ 7.0k
8d
In the same mouse-brain dataset, GPT-Rosalind found a possible clue to why TSC2 mutations tend to cause more severe tuberous sclerosis than TSC1 mutations.
@OpenAI
♥ 711
8d
Thu, Oct 22
6
I built a Command & Conquer style RTS called Battle Drones, mainly using Claude Opus 5.5, which runs in the browser and includes device-to-device multiplayer over an audio link
@simonw
♥ 762
9d
Has anyone seen the prompts used by these awful reply bots? I'd be fascinated to know how they make their posts so soulless and infuriating
@simonw
♥ 1.0k
9d
One of them replied!
@simonw
♥ 227
9d
Local AI 202: Get Massive Speed-ups with Speculative Decoding https://t.co/3ywucgYbrD
@huggingface
♥ 160
9d
https://t.co/BUVWDXEWXe
@GoogleAI
♥ 149
9d
Our latest model advances bring stronger capabilities for building and migrating software in low-resource programming languages such as COBOL, Fortran, and Visual Basic.
@OpenAI
♥ 9.5k
9d
Wed, Oct 21
2
Mark your calendars 📅
@conso_xyz
♥ 867
10d
Introducing Topview T1 🎬
@TopviewAIhq
♥ 354
10d
100
of 396 archived signals shown
·
agents get the rest over MCP
Living answers
For your agent
this workspace is one MCP call:
read(topic)
.
connect in one line →