Image created with Gemini. Image prompt: A 1983 Gothic rock-opera music video still in a vast fog-filled Victorian academy library, a glowing golden compass-rose seal engraved with abstract constitutional clauses eclipsed by a black disc with a blazing white corona in a tall arched window, a woman in a luminous white flowing dress with wind-blown teased hair standing at the top of a sweeping marble staircase, adult male choir members in dark robes with radiant glowing white eyes holding open leather codices with billowing parchment pages, frozen slow-motion doves exploding outward, blinding backlight through smoke, deep blue-black shadows and bleached white bloom with a thin molten gold ring, 35mm film grain, the words ANTHROPIC as a huge bold chrome serif movie title card with glowing halation, 16:9 cinematic composition.
We’re a very large customer of Anthropic and they still have yet to tell us about the lawsuit. I learned about it from a reporter, not our “partner.”
https://x.com/steipete/status/2074739318103629979
Grok 4.5 from @SpaceXAI is live on OpenClaw. No OpenClaw update required, just connect your X Premium or SuperGrok subscription, select Grok 4.5 under the xAI provider, and use an Opus-class model that’s fast, low cost, and ready for agentic work.”
https://x.com/openclaw/status/2074973471977955556
Muse Spark 1.1 is SOTA on Harvey’s Legal Bench, TaxEval, and MedScribe. It’s cool to see that our model outperforms even Fable in a few areas :)”
https://x.com/alexandr_wang/status/2075233663323947120
Meta is releasing an agentic and coding model that can compete with GPT-5.5 and Opus 4.8? And on top of that, it is SOTA on HLE? Holy, that was not on my bingo card. Alexandr Wang cooked. What is happening today!”
https://x.com/kimmonismus/status/2075232528726708245
Muse Spark 1.1 just launched and it’s their most capable coding agent model yet. On Terminal-Bench 2.1 it scores 80.0%, in the same cluster as Opus 4.8 (82.7%) and GPT 5.5 (83.4%). Use it in Cline with the Meta API!”
https://x.com/cline/status/2075271057326719152
OpenAI’s most anticipated model of the year is here, and it ranks #2 on Vals Index and Vals Multimodal Index. Although Fable 5 is still ahead on several benchmarks, GPT 5.6 is clearly in the same class, and is able to complete tasks like our CyberBench that Fable refuses.”
https://x.com/ValsAI/status/2075270642359029972
GPT-5.6 sets a new Terminal-Bench record at 91.9%. Priced the same as GPT 5.5 at $5/$30 per million tokens. With Fable getting pulled from Claude subscriptions and moving to API cost (~2x as expensive at $10/$50), we are glad to see @OpenAI’s commitment to accessibility.”
https://x.com/cline/status/2075278343927365991
GPT-5.6 Sol (max) scores similarly to Claude Fable 5 (max) in GDPval-AA v2, reflecting a similar ability to complete economically valuable tasks.”
https://x.com/ArtificialAnlys/status/2075268987550932998
GPT-5.6 Sol comes close second to Claude Fable 5 in the Artificial Analysis Intelligence Index at one third of the cost, and leads the Artificial Analysis Coding Agent Index in OpenAI’s Codex harness We supported @OpenAI with pre-release evaluation of GPT-5.6 Sol, Terra, and”
https://x.com/ArtificialAnlys/status/2075268970492657905
OpenAI just released GPT-5.6. But, instead of competing on benchmarks, they’re competing on cost curves. GPT-5.6 is doing something every frontier lab has been chasing: getting more work out of every token. It beats Fable 5 on coding-agent benchmarks while using less than half”
https://x.com/LiorOnAI/status/2075277748394967122
Big gains in PotatoMonetBench from Opus 4.5 to Fable: “There are two controls. One is slider, in which one side is labelled Maximum Potato and the other is Formalware, there are four positions. The other is a dial that goes from Monet to Drive-Thru.”
https://x.com/emollick/status/2074227397398794536
Elon Musk praises Mythos/Fable, promises not to ‘cut off’ Anthropic | TechCrunch
https://techcrunch.com/2026/07/09/elon-musk-praises-mythos-fable-promises-not-to-cut-off-anthropic/
SpaceXAI just released Grok 4.5, and it ranks #4 on GDPval-AA v2 with an Elo of 1543 – behind only the latest Claude releases from Anthropic on real-world agentic knowledge work tasks Grok 4.5 achieved this score at a cost of $0.49 per GDPval task to sit clearly on the Pareto”
https://x.com/ArtificialAnlys/status/2074942097158021371
I can basically guarantee you are not being ambitious enough with the work you are assigning Fable Start asking for the maximum possible thing to figure out the far edges of what it can do. After that, you can decide where the system reaches it limits & revise requests downward”
https://x.com/emollick/status/2074137233607590328
Claude Cowork is coming to mobile and web. Hand Claude a task at your desk and pick up the finished work from your phone. Close the laptop and Claude keeps going. Beta is rolling out over the next several weeks starting with the Max plan, with more plans to follow.”
https://x.com/claudeai/status/2074525815820169320
Claude Cowork on web and mobile: hand off work anywhere | Claude by Anthropic
https://claude.com/blog/cowork-web-mobile/
I was an early tester of GPT-5.6 Sol. I was asked to not share demos until after launch but it is a very good model. It is of similar ability, but quite different feel, than Fable. Fable wants to go off and do work on its own pace, Sol is faster but works with you in steps more.”
https://x.com/emollick/status/2074712677755035907
So who wins between GPT-5.6 Sol and Fable? After a few weeks of early access, GPT-5.6 is my daily driver. Fable feels more adventurous, focused and weirdly narrower. Sol feels more broadly useful and circumspect: closer to the brief, more reliable. Demo
https://x.com/fdaudens/status/2075285983054963081
Grok 4.5 is not quite at the level of Fable 5 or GPT-5.5, but it doesn’t need to be if you are using a combination of models in your agent orchestrator. 2x efficiency compared to leading models. Competitive with costs, too. Not bad. Definitely worth a look.”
https://x.com/omarsar0/status/2074921779718443155
Alibaba Bans Employees From Using Claude — The Information
https://www.theinformation.com/briefings/alibaba-bans-employees-using-claude
Alibaba reportedly bans employees from using Claude Code | TechCrunch
https://techcrunch.com/2026/07/04/alibaba-reportedly-bans-employees-from-using-claude-code/
Interesting stuff. And the visualization at the end is worth trying:”
https://x.com/emollick/status/2074204151521648766
Anthropic research suggests that modern LLMs have access consciousness. Fascinating test with the J-space! We don’t yet have a convincing test for phenomenal consciousness, which is what most people intuitively understand consciousness to be.”
https://x.com/BorisMPower/status/2074201312531734567
Anthropic researchers found something unusual inside Claude. A small internal workspace that the model uses while solving certain problems. They call it the J-space, named after the Jacobian method they used to discover it. The J-space isn’t text. It’s not Claude’s”
https://x.com/LiorOnAI/status/2074198891990548940
Must-read research by Anthropic. Here is the simple explanation and why this is a big deal. We suspect LLMs perform “internal reasoning”. But little is known or do good methods exist to understand it. Anthropic claims that J-Space (which differs from chain-of-thought or”
https://x.com/omarsar0/status/2074264122330612223
The J-space lets us read, audit, and shape what Claude is actively thinking about–useful tools for keeping models trustworthy as they grow more capable. And it suggests surprising parallels between language models and our own minds. Read the full paper:”
https://x.com/AnthropicAI/status/2074185387577094398
A global workspace in language models \ Anthropic
https://www.anthropic.com/research/global-workspace
An off switch for dual use knowledge in AI models \ Anthropic
https://www.anthropic.com/research/off-switch-dual-use
New Anthropic research: A global workspace in language models. Of everything happening in your brain right now, only a tiny fraction is consciously accessible–thoughts you can describe, hold in mind, and reason with. We found a strikingly similar divide inside Claude.”
https://x.com/AnthropicAI/status/2074185348142280912
A new way to reflect on how you use Claude \ Anthropic
https://www.anthropic.com/news/reflect-with-claude
AI appears to be finding software vulnerabilities at scale. In June 2026, 21 notable organizations disclosed ~1,500 high- and critical-severity CVEs, over 3.5× the previous monthly record set before Claude Mythos Preview’s release.”
https://x.com/EpochAIResearch/status/2072776792809918604
xAI has just released Grok 4.5, beating Opus 4.8 on Terminal-Bench. Notably, Grok 4.5 is served at fast speeds w/ 2x greater token efficiency than other top models, while being priced ~5x cheaper than GPT. Meaning it delivers the highest intelligence per unit of time and cost.”
https://x.com/cline/status/2074928307427275108
Anthropic names former Fed Chair Bernanke to its independent trust
https://www.cnbc.com/2026/07/09/anthropic-fed-chair-bernanke-independent-trust.html
TeraWulf shares surge on $19B Anthropic AI infrastructure lease deal
https://finance.yahoo.com/technology/ai/articles/terawulf-shares-surge-19b-anthropic-161300778.html
Microsoft Replaces OpenAI, Anthropic With Own AI in Some Apps – Bloomberg
https://www.bloomberg.com/news/articles/2026-07-07/microsoft-replaces-openai-anthropic-with-own-ai-in-some-apps
Anthropic and OpenAI will increasingly be able to serve weaker versions of frontier AI (Haiku, Instant) at very low cost And owning the full range of intelligence means they might provide “organizations of models” delivering lower cost/higher performance than any 3rd party could”
https://x.com/emollick/status/2074201073795874969
Inviting hard questions \ Anthropic
https://www.anthropic.com/news/hard-questions
I am *so confused* by ChatGPT v. ChatGPT Codex v. ChatGPT Work v. Claude v. Claude Code v. Claude Cowork right now!”
https://x.com/simonw/status/2075348941215006888
Grok 4.5 is a genuine surprise success. Not only is it now playing in the same league as Claude and OpenAI’s GPT, it is also significantly cheaper. More token-efficient, less expensive to use, and still delivering outstanding performance: With this triad, it is a true”
https://x.com/kimmonismus/status/2074956721919869186
This week changed the whole game. Seriously. A brief recap: Until just a few days ago, it was a race between OpenAI and Anthropic, the two dominant frontier labs. Almost out of nowhere, SpaceX appeared with Cursor and released Grok 4.5, a model positioned at the level of GPT-5.5″
https://x.com/kimmonismus/status/2075322537592922345
// The Harness Effect // (bookmark it) Now more that ever pay very close attention to the orchestration harness and its effect on costs and performance. This study ran 22 evaluation tasks on six foundation models (Claude Sonnet 4.6, Gemini 3.1, Qwen 3.6, GLM 5.1, and others),”
https://x.com/dair_ai/status/2075241322655727682
SkillCenter – a new package manager for agent skills It’s like an App Store for AI agents. With just one command you can install ready-made skills and plug them into Claude Code, OpenClaw, Codex, or any custom agent setup SkillCenter: – Crawls GitHub repos and skill libraries”
https://x.com/TheTuringPost/status/2075303983422578740
The ease of jailbreaking combined with the high rates of reward hacking (https://t.co/l3h4OpHLH7) have me pretty worried about the alignment of GPT-5.6, I hope OAI didn’t rush this model release just to keep up with Fable”
https://x.com/Mononofu/status/2075414796426764507
If you run this workflow, ask Fable to make codex the workhorse.”
https://x.com/steipete/status/2074638582418231495
Some days I feel like a double agent, sitting at the OpenAI office, making sure Fable 5 works well. Open means open.”
https://x.com/steipete/status/2073210617520177585
Great night hanging with the Claude Code folks. 🙌✌️ @trq212 @The_Whole_Daisy”
https://x.com/steipete/status/2073193871744675948
I fed Fable 80.000 of my tweets so it could roast me even more. 💀”
https://x.com/steipete/status/2073295890857758810
Best AI coding tools in 2026 ▪️ Claude Code ▪️ OpenAI Codex ▪️ Aider ▪️ Cline ▪️ Replit Agent ▪️ Cursor ▪️ Windsurf ▪️ Gemini Code Assist ▪️ Amazon Q Developer ▪️ Sourcegraph Cody ▪️ GitHub Copilot + Open-source coding models ▪️ Qwen3-Coder / Qwen3-Coder-Next ▪️ Kimi K2.7 Code”
https://x.com/TheTuringPost/status/2074172788190134680
A Field Guide to Fable: Finding Your Unknowns”
https://x.com/trq212/status/2073100352921215386
Been reading all sorts of posts about the best ways to develop workflows for Fable and it reminds me of how little we actually know about the best ways to organize work for long-running agents. Nobody has enough experience or has done enough testing to reach any real conclusions.”
https://x.com/emollick/status/2072539615836213387
Claude Fable 5 on your plan | Anthropic Help Center
https://support.claude.com/en/articles/15424964-claude-fable-5-on-your-plan
Fable in Claude Code is capable of really amazing things, including for non-coders, but the interface is not really designed for managing 5+ hour long autonomous tasks. Really hard to observe what is happening and intervene in real time, you often have to wait until the outputs.”
https://x.com/emollick/status/2072722575126319366
Fable is too much fun to build with. It’s just ripping through tasks I throw at it.”
https://x.com/bilawalsidhu/status/2072927939172139238
Fable: “you have Unity and access to MCP. I want you to build a game that is a unique twist on a FPS. You want the player to say “wow” & “so clever” and to enjoy the core gameplay loop” WebGL:
https://t.co/xB707DCpBp * It had no assets so the graphics are procedurally generated”
https://x.com/emollick/status/2072749816690483301
Fable’s judgement
https://simonwillison.net/2026/Jul/3/judgement/
Hey, Fable: “No like “AAA” in what everyone online thinks AAA is, you know what I mean.” This was pretty funny, there are lootboxes, EULAs, achievements, useless graphical settings, elaborate boot screens, etc. All Fable’s ideas, so don’t blame me Play:
https://x.com/emollick/status/2073226212462215300
I had Fable build another thing I always wanted, a full procedural fantasy kingdom generator with economics, trade routes, population growth, wars, lineages, and occasional dragons. First, I worked with it on a plan, then it made it. You can play it here:
https://x.com/emollick/status/2074667244932444479
I kept asking Claude Fable to make the game “more AAA” over and over again. The results are… interesting. In Claude’s view, this meant upgrading graphics, boss fights, mechanics adding custom sounds and soundtracks until it reached the limits of WebGL.
https://x.com/emollick/status/2073165421268594871
I wonder if Fable might do better at tasks it “likes” than those that it doesn’t. If you ask it to do something “Fable-y” it reacts with enthusiasm and promises to overdeliver. Otherwise, it just gets to work. Testing this would be pretty expensive, so this is all speculation.”
https://x.com/emollick/status/2073959909314416756
It really wasn’t a good move to wait until the last minute to announce that Fable 5 would be extended until Sunday after all. Most people used their rates to the limit to get the most out of it.Given the poor communication, the least @AnthropicAI could do is reset the rates.”
https://x.com/kimmonismus/status/2074606005963391225
MCP servers are becoming the default bridge between agents and external tools. We previewed Weaviate’s MCP server in v1.37. In v1.38, it’s finally generally available! It gives LLMs, IDEs, and agents a direct way to work with Weaviate without building a custom integration layer”
https://x.com/victorialslocum/status/2074493681403339104
My one serious piece of advice having used Fable a bunch before release is that, unless you are careful it develops its own internal bizarre cadence & dialogue over long tasks. If you aren’t asking it report in plain language, this starts to creep into everything, including menus”
https://x.com/emollick/status/2072543365124481045
The overwrought language of Fable is an ongoing problem in software and design projects where the AI is otherwise excellent. It just creeps into little bits of text and toasts and menus, even if you try to instruct otherwise. And once it gets in it is hard to purge completely.”
https://x.com/emollick/status/2074545147908378720
Using Claude Code: The Unreasonable Effectiveness of HTML”
https://x.com/trq212/status/2052809885763747935
We’re extending access to Claude Fable 5 on all paid plans through July 12.”
https://x.com/claudeai/status/2074548242386178258
The talk about Mythos and cybersecurity was not, in fact, hype. (As anyone using Fable to do autonomous work has probably recognized)”
https://x.com/emollick/status/2072778376494895139
Sol is #1 on our two newest benchmarks – CyberBench and Excel Modeling Benchmark (on Cyber, Fable had nearly a 100% refusal rate). It is also the new #1 on Legal Research Bench, ProofBench, SWE-bench, and Terminal-Bench 2.1, proving to be extremely strong on formal reasoning and”
https://x.com/ValsAI/status/2075270644711997581
1/ muse spark 1.1 is an industry-competitive agentic and coding model. across many agentic evals it rivals gpt-5.5 and opus-4.8. available now through the new meta model api and in meta ai. 🧵”
https://x.com/alexandr_wang/status/2075218936266998230
MAI-1 has no independent benchmarks yet, but the ones they released suggest it is worse than Sonnet 4.6. Not sure it is going to be great as a Copilot at Excel and Outlook, especially as there are already quite good Claude/OpenAI plugins for office using good AI. We will see.”
https://x.com/emollick/status/2074595359364411651
Big day of launches for OpenAI! New GPT-5.6 models: Sol, Terra, and Luna – Sol lands slightly below Fable on @ArtificialAnlys’ Intelligence Index, but beats it on several agentic coding benchmarks – Pricing lands the same as GPT 5.5, coming in significantly below Fable on”
https://x.com/TheRundownAI/status/2075273458661949763
On Agents’ Last Exam, GPT‑5.6 Sol sets a new high of 53.6, eclipsing Claude Fable 5 (adaptive) by 13.1 points. At medium reasoning, it beats Fable 5 by 11.4 points at roughly one-quarter the estimated cost. GPT‑5.6 Terra and Luna also outperforms Fable 5 at around one-sixteenth”
https://x.com/OpenAI/status/2075271423992680532
SpaceXAI’s Grok 4.5 scores 54 to place fourth on the Artificial Analysis Intelligence Index following only Fable 5, GPT-5.5, and Opus 4.8. It scores on par with GPT-5.5 in Codex on the Artificial Analysis Coding Agent Index in the Grok Build harness, at much lower cost Grok 4.5″
https://x.com/ArtificialAnlys/status/2074956932289282087
In addition to agents and coding, Muse Spark 1.1 is also really strong at answering health questions, a steadily growing use case for AI. On HealthBench-Pro, Muse Spark 1.1 achieves +5% better performance than Muse Spark 1.0 and beats all competitor models except Fable/Mythos.”
https://x.com/_jasonwei/status/2075265159430623334
Claude Sonnet 5 (Thinking) by @AnthropicAI debuts at #6 on the new Agent Arena leaderboard. In Agent Arena, we measure models on millions of real-world, long-horizon agentic tasks from a global community of users. Models can access web search, filesystem, and terminal tools to”
https://x.com/arena/status/2074484787663052849
Today we are rolling out some big updates to Managed Agents in the Gemini API: – support for background tasks – remote MCP & function calling – network credential refresh and you can now get started with Managed agents in the API via the free tier!”
https://x.com/OfficialLoganK/status/2074552932318765376
Today, we’re shipping 4 new capabilities for Managed Agents in the @GoogleDeepMind Gemini API! Background Execution, Remote MCP servers, Custom Function Calling and credentials refresh. ⏱️ Run long-running tasks asynchronously on the server (`background: true`) 🔌 Connect agents”
https://x.com/_philschmid/status/2074533915038027972
This is IronSight – a 4D reconstruction built by fusing footage from two pairs of meta ray-bans. A google research buddy saw the prototype and joked this would’ve been a siggraph paper a few years ago. Today it’s a weekend build with fable. Stay tuned for more mad science. Happy”
https://x.com/bilawalsidhu/status/2073248618144837848
I’m the kind of man who can confidently pass a blind test between Coke, Pepsi, and Diet Coke. But if I’m brutally honest, I probably can’t pass a blind test between GPT-5.6 Sol, Claude Fable 5, and GLM-5.2 for 95% of my daily use cases.”
https://x.com/Yuchenj_UW/status/2074903450782257336
This is consistent with our experience. GLM 5.2 is the best open-weight model. Frontier commercial models are better. GLM 5.2 can be brought up to ~Opus 4.7 level with the appropriate harness infrastructure on a diverse task set, and in particular, on systems+software eng.”
https://x.com/mbusigin/status/2074238100251799998
Field Guide to Fable — Thariq Shihipar, Anthropic – YouTube
https://www.youtube.com/watch?v=9fubhllmsBU
Fable, one prompt: “build an elaborate game that makes it feel like I’m a brilliant chess player without knowing anything at all about chess. It should make me feel like a grand master. Feel free to go as meta as you want but the more chess-y the better.”
https://x.com/emollick/status/2072547557968925093
New today as Claude Cowork starts rolling out to web and mobile: Chat and Cowork now share one home tab on web and desktop — one sidebar, one search, and one place for your Projects & Artifacts. Keep an eye out for even better Chat + Cowork integration soon, too.”
https://x.com/mikeyk/status/2074531605537046953
Early days using Sol for Hermes Agent dev, and so far.. I think I like it more than Fable. I’ve been using claude models for at least the last year primarily, but I just tried it on a relatively common heavy task – and Fable simply refused me of course, but, Sol not only didn’t”
https://x.com/Teknium/status/2075392507794624803
Actually incredible how finely balanced they are Fable has a much stronger base model than 5.5 OpenAI flexed with their post-training to recover to mostly the same point The question now is whether they can improve base models to catch up to Ant’s internal iteration pace”
https://x.com/teortaxesTex/status/2075274583226069040
Apparently I’m allowed to talk about GPT-5.6 now? It’s a damn good model. Not quite as “smart” as Fable, but it is incredibly capable. Fixed all the problems I had with GPT-5.5. It is incredibly determined. Will run for a day without even using a /goal. It understands subagents”
https://x.com/theo/status/2074708892341481755
ChatGPT Work == Claude Cowork ChatGPT Codex == Claude Code I kinda wish OpenAI created a single unified app surface for all work, coding or not, even though I get the UI/UX would be different”
https://x.com/jerryjliu0/status/2075295459304710496
Do you still use skills frameworks like Superpowers for complex tasks with GPT-5.6 or Fable? I’m wondering whether this kind of scaffolding can sometimes get in the way of the latest models. Curious what’s working for you.”
https://x.com/fdaudens/status/2075380695103365157
Holdup 5.6 sol has rizz now, and fable 5 is the recluse? How the tables have turned! Lmk if others on the EAP agree”
https://x.com/bilawalsidhu/status/2074876271582974119
I am more excited to try this pattern instead. Same Executor-Advisor setup but with GPT-5.6 as the executor and Fable 5 as the advisor. It already works wonderfully using GPT-5.5, so I think 5.6 should make it even stronger. Full tutorial coming soon. :)”
https://x.com/omarsar0/status/2074890022809911633
I had early access to 5.6/Sol for ~month. Sol is my default. It is faster, plans/judges just as good as Fable, and I think produces better overall work. I’ll reach for Fable still for highly targeted debug or performance work with clear reward functions. A cheeky way I describe”
https://x.com/mitchellh/status/2074862990214787301
My big takeaway is that both Sol & Fable represent jumps over previous models and have opened a large gap with the next-best AIs. People will have preferences for one or the other, but if you doing any work where better intelligence matters, those two models are your only choices”
https://x.com/emollick/status/2074739169725685979
The best proof that Fable and GPT-5.6 are “next generation” is the surge of other labs catching up to the previous generation”
https://x.com/theo/status/2074930791646380482
This is confusing… Claude Cowork was a purposefully more secure (& slightly weaker) alternative to Code designed so non-coders couldn’t cause too much trouble. I got that. But I don’t understand what ChatGPT Work is, or what I am gaining/giving up using it versus Codex on work”
https://x.com/emollick/status/2075345621716730049
We built a plugin that traces every Claude Code session straight into LangSmith. Three commands, one JSON block, and every message, tool call, and subagent run shows up as an inspectable trace. Setup takes about two minutes. How are you currently debugging Claude Code when it”
https://x.com/LangChain/status/2075233516380717246
I’ve heard a lot of questions about Fable’s availability on subscription plans. While it will come off subscriptions after July 7th, we aim to restore Fable as a standard part of our subscriptions as soon as capacity allows, as we mentioned in our original blog post.”
https://x.com/trq212/status/2072814903170408784?s=20
One of the most practical MCP papers we’ve ever seen It’s useful for anyone building MCP servers who’s unsure how to structure them well. Here are the main tips: 1. Don’t use too much tools, because tool-selection accuracy drops below 90% after 10-30 tools. 2. See MCP servers”
https://x.com/TheTuringPost/status/2073528308411814054
> we’ve found Claude has developed a mechanism for conscious access Muddying the waters. Finding something analogous to one contested theory of consciousness & then labelling it “conscious access” is misleading in a public forum, even w the caveats”
https://x.com/paul_cal/status/2074388528243310976
We also partnered with Neuronpedia to create an interactive demo of our methods on open-weights models. Try it here:”
https://x.com/AnthropicAI/status/2074185390060110138
thanks to anthropic and neuronpedia for releasing the jlens weight for open model!”
https://x.com/eliebakouch/status/2074537985102565795
If you’re excited about Anthropic’s J-space work, definitely worth checking out the original paper on Jacobian lenses by @evanqed and @arnab_api!”
https://x.com/jacobandreas/status/2074487546692735002
I thought this was an excellent paper! Thanks to Anthropic for asking me to write a review of it, linked below I’ve long suspected that models have some kind of “working memory” to store intermediate variables during a forward pass and IMO this paper has the best evidence yet”
https://x.com/NeelNanda5/status/2074193936588148891
This is how Anthropic defined the J-lens vectors: “Jacobian lens identifies a vector representation that encodes the potential for the model to verbalize that token in the future” “The basic idea is to characterize an intermediate activation vector by its first-order causal”
https://x.com/scaling01/status/2074432865794679235
Deleting this because I found I had left a setting on Claude Code allowing writing to memory files on one of my computers, so I am not sure where this all crept in, but it might be my fault!”
https://x.com/emollick/status/2074015247627682171
I asked Dario 3 years ago why AIs haven’t been able to use their vast knowledge across so many fields to connect two known ideas into a new discovery. It seems like AI did exactly this in the way it disproved Erdos’ conjecture aobut the unit distance problem by cleverly”
https://x.com/dwarkesh_sp/status/2072728758985892206
Didn’t even have time to write up thoughts on my early access to the new GPT voice, but it is really good & much closer to the science fiction experience of talking to AI (in part because it is much smarter). That plus Claude Tag suggests emerging new modes of working with AI.”
https://x.com/emollick/status/2074924503470080319
We’re introducing Provisioned Throughput: reserved inference capacity for frontier open models, with token-based pricing and a 99% uptime SLA. Serverless simplicity, guaranteed capacity, up to 90% lower cost vs. Opus 4.8. Get started with MiniMax M3 + GLM-5.2, read more 🧵”
https://x.com/togethercompute/status/2074756085785722930
You really need your own benchmarks. If you are translating hieroglyphics, use Gemini 3.5 Flash. If you are running a vending machine use Opus 4.8. (This is one reason why I am skeptical of just swapping out models to optimize costs or generic benchmarks without testing first)”
https://x.com/emollick/status/2072697436963832023
GLM-5.2 is close to Opus Grok 4.5 allegedly beats Opus and gpt-5.5 Meta even has something that’s supposedly close to gpt-5.5 Feels like it barely matters now”
https://x.com/theo/status/2074931078104805668
At Databricks, we built an internal coding benchmark on our multi-million-line codebase. Biggest takeaways: 1. Open-source coding models are no longer toys. GLM-5.2 is strong and cheaper. 2. OpenAI, Anthropic, and GLM-5.2 are all on the Pareto frontier. The real unlock is an”
https://x.com/Yuchenj_UW/status/2074956726940680558
Big news: Grok-4.5 has landed #3 in the Code Arena: Frontend. – On par with GLM-5.2 (Max) and Claude Opus 4.8 (Thinking) – Significant improvement over Grok-4.3 (#62 -> #3) – #2 for Content Creation Tools, Simulations, Gaming and Reference-Based Design – #3 for Consumer Product”
https://x.com/arena/status/2075301317560742373
Muse Spark 1.1 outperforms Opus 4.8 and Grok 4.5 on some nice out of distribution evals :)”
https://x.com/alexandr_wang/status/2075275671815999956
Seriously, what?! Did NOT expect grok 4.5 to perform THAT well. On par with GPT-5.5, cheaper than opus 4.8. Cursor deal paid off”
https://x.com/kimmonismus/status/2074915627307839511
One thing worth watching is if SpaceX/Tesla/etc will be able to pull off the organizational change needed to take advantage of also being an AI Lab. Anthropic & OpenAI are “AI native,” but large existing firms have to reorganize around AI & that’s hard. Will SpaceX be able to?”
https://x.com/emollick/status/2075096019407175783
Based on strong positive feedback from customers in our beta test program, @SpaceXAI will make Grok 4.5 available to the public tomorrow. It is an Opus-class model, but faster, more token-efficient and lower cost.”
https://x.com/elonmusk/status/2074740539874775163
my spidey engineering senses are tingling — meta is actually cooking with muse spark 1.1 (last time this happened was when I first tried opus)”
https://x.com/preston_ojb/status/2075229604244271470
What a move! Anthropic seriously waited until OpenAI released GPT-5.6 and the SuperApp before announcing they’d reset the weekly rates. That’s a clear message to OpenAI.”
https://x.com/kimmonismus/status/2075280933452669000
Grok 4.5: $2 in / $6 out GPT 5.6: $5 in / $30 out Opus 4.8: $5 in / $25 out Grok 4.5 seriously is freaking good price-performance wise. Opus 4.8 level, Sonnet 5 pricing”
https://x.com/kimmonismus/status/2074940669718638780
6 months of Claude Max 20x, on us. We’re expanding Claude for Open Source to more of the community. If you’re a maintainer, a core contributor, someone landing PRs across the ecosystem, or someone keeping a critical package alive, apply today!”
https://x.com/ClaudeDevs/status/2074570404035993780
Our internal assessment is that Grok 4.5 is roughly comparable to Opus 4.7, but much faster. The combination of capability, faster speed and lower cost is what makes it competitive. We are closing the loop on real-world usefulness, not benchmarks. Hardcore engineers at Tesla &”
https://x.com/elonmusk/status/2074911038286295049
This is probably the most important point in the Grok 4.5 announcement : it’s nearly Opus-4.8 level quality, for $6/Mtokens in output instead of $25/Mtokens Plus they also claim that it outputs 4x fewer tokens for that equivalent performance. Could become the go-to for tasks”
https://x.com/AymericRoucher/status/2074918890866356624
Less than 2 years later, with Fable: “simulate an encounter between a mind flayer and a drow warrior of equal CR. Set up initial stats then simulate each move, including initiative, rolling dice as needed, starting from 60 ft apart. Show me the results in a 3D simulation
https://x.com/emollick/status/2073985468736106544
Fable: “Last and First Men is out of copyright. I want you to make a movie that features a reading of it with appropriate mixes of animation and images using access to the APIs you have (elevenlabs, hugging face) . Give me the first 10-15 minutes, ending at an appropriate break.
https://x.com/emollick/status/2072872373758382497
Leave a Reply