bitmin

@johnennis Have, but Grok isn't doing it for me. I know.

Here is how my AI subscriptions are shaking out: Codex (obviously) - $200 Claude Code (for now, it's love/hate) - $200 Manus (literature review and research) - $200 Sakana Fugu (critical review and knowledge synthesis) - $200 Grok Build (generic worker plus image/video) - $99 I'm also flirting with Devin, it's insanely fast and could become a second pair of hands, but the harness is not as mature so it doesn't fit into my workflow as easily I go back and forth on Cursor Composer, it's very good, but Grok Build pretty much does the same thing plus you get the image/video Any other recommendations?

@AlexFinn Also, 8. Be flexible? Here's Fable 5: "The strongest audit is a decorrelated set of eyes: Kimi (different model with a wider net), and an independent Fable-5 (verify + seam-hunt between the chunked modules)."

The most dangerous thing you can do right now is NOT use the latest AI tools. Period. Every day a new company is laying off thousands of people who don't know how to use the most modern AI tools If I were in the 9-5 world right now, this is every step I'd take: 1. Download Codex and build your first app. Learn how to implement a front end and database (I’d use nextJS and Convex). AI can teach you all of this 2. Download Hermes agent. Tell the agent about your entire life. Career, goals, and ambitions. Ask it what workflows it can implement to get you closer to those goals. This is called reverse prompting. 3. Build your own Mission Control with Hermes agent. This is your own custom software that automates every part of your life. Anytime you do a task on your todo list manually, give it to Hermes and figure out how you can automate it in your Mission Control 4. Get the Claude Max plan. Feed Fable 5 Max your hardest problems. Make sure you hit the rate limit on it EVERY single week 5. Constantly look at your limits in all your AI plans. If you're ever not above 50% on your limits, get angry that you're not burning enough tokens. 6. Learn how to use Claude Code side by side with Codex. Learn both their strengths. Fable 5 for planning, ChatGPT 5.6 sol medium for execution. These are the only 2 coding tools that matter. Master these and you're golden 7. Start creating content and building a platform. Resumes no longer matter. Companies dont look at them anymore. They check your platform. What you talk about, what you’re passionate about, how you educate others, who follows you. This matters 100x more than anything else when getting hired right now. If you do these 7 things you are in excellent position to not only be safe in your career, but also dominate those that don't pick up these skills.

@APompliano The West won’t win. China’s open-weight models are a direct response to Western protectionism—cheaper, government-subsidized, electric bills at 1/10th the cost and only months behind. Super curious what the next few years will bring. Normal people are expecting an imminent ban.

Banning Chinese models is the most un-American thing we could do. Stop complaining and start competing. People won't use Chinese models if American companies build something better.

@0xileri Galloway has been right about this for more than a year. He called it, he is right, the charts from OpenRouter are the proof.

Business professor Scott Galloway says China is flooding the world with dirt-cheap AI models on purpose, and it’s wrecking American AI companies faster than cheap Japanese cars ever wrecked Detroit. “Free Chinese models went from less than a third of all traffic in late 2025 to about two-thirds recently.” “They steal the IP on a cell phone tower from Siemens, and then they manufacture it with cheaper labor and they sell it back for 40 cents on the dollar.” “The best business model in history is IP theft.” “What the Chinese automobile industry did to Detroit in about three decades, it feels like Beijing is doing to us in AI frontier models in about three months.”

@leo_linsky Kimi is the only LLM I've been able to use for auditing because of it's big context window. Fable 5 has to chunk and worse, it loses essential nuance between the chunks.

Kimi K3 results are now live on GBENCH: We have a new frontier open weights model in agentic coding, by an immense margin. We ran it through coding evaluations in complex games where it directly competed or cooperated with code agents written by other models. It truly is a frontier model, surpassed only by Sol and Fable. It designed smarter programs than Opus 4.8 and GPT 5.5. It's worth noting that launch day inference is about 3x slower, and it's not as capable in one-shot fluid intelligence. Still, an absolutely amazing release by @Kimi_Moonshot. Frontier-level intelligence, soon to be openly released to everyone. The singularity is ramping up.

@VaibhavSisinty <Attach your source of truth here = "SOURCE">

A college student just killed the biggest advantage politicians have ever had. The gap between the lie and the fact-check. He built a Chrome extension called InTruth. It listens to any live speech, debate, or interview. Transcribes every word. And checks every claim against real sources in real time. Before the speaker even finishes the sentence. It works on debates. Press conferences. Town halls. Interviews. Anything with someone talking on video. The fact-check shows up on your screen while they're still talking. Until now, fact-checking happened hours or days later. By then the clip has been shared, the headline has been written, and the damage is done. Nobody reads the correction. InTruth closes that gap to seconds. The claim and the fact-check exist in the same moment. A college student, a Chrome extension, and an AI model just did what entire newsrooms couldn't figure out for decades.

@LuminaXspace That race is petering out and it's becoming cost / tokens...

🚨Z.ai may be skipping GLM-5.3 and going straight to GLM-5.5 • Expected to launch in August • Potentially more than 1 trillion total parameters • Will likely remain Open Weight • Should retain its 1M token Context Window • Built for long running autonomous coding and agent tasks This model should blow Kimi K3 and Qwen 3.8 out of the water. Do you think it will close the gap finally and challenge Anthropic/OpenAI?

@bridgemindai Using it, but knowing in my heart they are using my code to iteratively train Kimi without my permission. I get one-off good results, they get their K4 and the end of capitalism's big dreams of the future. 🫤

Kimi K3 is running 100% FASTER than it was this morning. Why? Moonshot hit capacity and chose to stop selling NEW subscriptions instead of throttling existing ones. Every plan: sold out. On purpose. When Anthropic hit the same wall in April, they cut existing users' usage 50% during peak hours. Moonshot cut their revenue. Anthropic cut your usage. That says everything. My Kimi K3 subscription is not going anywhere.

@bridgemindai It's an audit

@deanwball You will be proven "right" only because Chinese LLMs will be banned in the West - and soon. Otherwise, not right. Intellectual property theft? Maybe. China is advancing the state of the art. Looks like the trillions in Western AI is going the same way as EVs?

Some observations on Kimi: 1. It's a very good model! I don't think its performance can be explained away by distillation or anything like that. In agentic coding sessions, it seems pretty much on par with the best public models of Q1 2026. In my fairly limited use, it also seemed very token hungry. It's not obvious to me that this model is actually that cheap to run. 2. I am personally surprised the Chinese state continues to allow the open sourcing of models this good, given potential risks. To be clear, I *myself* might be fine with models presenting this level of marginal risk being open weight, but I am surprised that China is fine with it. I suspect the reason they are is 75% explained by strategic blindness/lack of AGI-pilledness (the CCP is very Yann Lecun-y in its views of AI). The other 25% or so is their lack of compute for customer inference (making China's open-weight strategy an unintended byproduct of US export controls) and the normal Chinese strategy of aggressive exports. For the companies, as opposed to the government, the decision to open source is partially ideological and partially because they are behind, and they know that very few people would pay for sub-frontier models from China. 3. Open-weight models are inherently decelerationist, and I'm continually surprised to see the so-called "accelerationists" so excited about open-weight models. I suspect the reason they are is that they know open-weight models are effectively ungovernable, and they simply like the overall cloak of ungovernability open-weight models create over the whole of AI. It's not a bad strategy; it reminds me of James Scott's recounting of the hill people in "the art of not being governed." Still, in the end, open-weight models deter further AI capex. 4. One probable outcome of an open-weight-model-dominant world is full AI communism, which is precisely what China proposes: rather than a market product, AI is a "public good" which will ultimately be provided by the state as a kind of "digital public infrastructure." This future strikes me as a dystopian hellscape, but I've never met an open-weight models advocate who doesn't ultimately concede this is where things end. You'd be surprised how many 'accelerationists' lobbied me, while I was in government, to support an eleven or twelve-figure federally funded data center so that startups could train models at a subsidy and then give them away for free. There was no other way for AI to progress, they said. Perhaps this is the logical end state of things. Nonetheless, I find myself surprised to see supposed accelerationists excited about such an outcome. I think many of them just don't know what they're doing. Many accelerationists do not view the creation and serving of frontier models as a legitimate business. 5. I would guess that the Trump Administration will at some point realize that their best strategy here would be to create large amounts of regulatory risk around the use of open-weight Chinese models. You don't need to "ban open source" (one of the dumber motifs of AI policy discussion). You just need to direct every agency to issue soft law that creates FUD. "A Federal Reserve Advisory Bulletin found that there may be backdoors in Chinese AI models." It needn't be that well justified. You just create enough regulatory risk that every regulated enterprise backs off. You probably don't want to create so much regulatory risk that you scare off the hyperscalers from serving Chinese models; this will just drive startups to sketchier providers. There's a happy middle ground here. I'd assume they will do some version of this. 6. It's probably true that open-weight models of this capability make the world a bit more dangerous, but not so much more that you'll really notice. At some point the models will be capable enough that you will notice. "A nonliving, invisible, dangerous, and infinitely self-replicating agent escaped from a Chinese lab," you say? Color me shocked.

@bridgemindai My controversial take is that Chinese LLMs are the next thing to get banned or tarriffed. Or maybe that isn't controversial at all. I presume Elon has already made his call and there is an executive order being drafted. Get your coding done while you can.

@X_is_Arbitrary @enzo_gte Not

@enzo_gte So the Chinese EVs are so good yet Tesla is the top selling electric car manufacturer in China.

I've been using Kimi K3 for ~16 hours now. The model is clearly good at a lot of different things (especially frontend), but non obvious reason why people are enjoying it so much is that it clearly does not follow the same rules in terms of safeguards and copyright. Kimi will happily clone MacOSX. If you ask it to help you improve another AI model, it will do it with a smile on its virtual face. Ask Fable to do the same thing? It literally starts to perceive you as a criminal committing a war crime (like no bro, all I want to do is fine tune an open source model). After using all three recent releases, Fable, GPT 5.6, and now Kimi, it's clear that the full power of the models has been significantly held back by the safeguard restrictions caused by last months debacle with the USG -- leading to the top models being quite literally lobotomized in some areas, which leads to subpar results as the safeguards pollute its entire thinking and problem solving abilities. The funny part? Is that you could have predicted this outcome 2-3 years ago when you started to see the rise of Chinese EVs and smartphones compared to western alternatives. They quite literally tried to copy the Tesla Model S and iPhone as hard as possible and then eventually it started to diverge to the point where their EVs and phones are just genuinely better (which is why we have export controls banning their EVs, because they would literally drive all US manufacturers to ZERO) There is a very clear behavior difference in Chinese capitalism and American capitalism. American capitalism tries to protects copyright, patents, etc (oh no, you can't download a book through LibGen, that's ILLEGAL!). Versus Chinese capitalism actually just does not give a fuck. "Hey you want a video gen model (Seeddance 2.5) trained on every single anime ever? And you want the main character to look exactly like Messi? Sure, here you go!" You see what I mean? When one half of the competition is being held up by regulators and restrictions on people who don't understand the technology and the other half has a leader who quite literally today said they are going to set up AI centers around the world to help other countries onboard to their open-source AIs, this is the sort of results that you will start to get. These models were not smart enough to have this difference in philosophy matter -- but the newest class of models is where this difference makes a big deal. If these models are finally at the point where they are smarter than 99% of humans, why would you want to use the American one who tries to impose its world view onto you versus the Chinese one who will just do what you say without asking any questions? And this isn't a full on bullpost on Kimi, the model is clearly not as smart as Fable / GPT 5.6 on things like math and science, but it's lack of handcuffs means that it can show the world what the frontier labs are gatekeeping from you and that starts to build customer resentment and loyalty towards the East, which is probably not what the USG wants. Interesting times. Interesting times, indeed.

@enzo_gte Next ban will prevent and then sanction anyone using a Chinese model. "After legal inquiry, and Moonshot's failure to reply, we have applied sanctions to the company and will block use." It's coming.

Chinese capitalism also doesn't give a fuck when your EV crashes and it turns out their safety protocols are: send a letter in Chinese to a PO Box and we will send you $30 for your deceased spouse. Their hoverboards catch on fire so frequently, you can't charge them in most places

@elonmusk When Opus 4.9

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 & SpaceX find Grok 4.5 genuinely useful, which is what actually matters.

@NadelParis @SawyerMerritt Elon is being 'had'. 😂. I think Elon is running the simulation.

@SawyerMerritt The children will likely never see a penny. Trump and his family don’t do charity. Elon is being had as usual. His executives are being screwed as well. Trump and his family are all scumbags.

SpaceX President and COO Gwynne Shotwell and her husband have announced that they are donating ~$325 million of SpaceX stock to Trump Accounts for more than two million children across the U.S. Every one of those kids will now have a direct stake in a company whose mission is the most ambitious in human history: to make life multi-plantary 🚀

@Bencera The goal posts are moving so quickly. Decent chance China even exceeds the West's inference soon... Then USA will ban use.

China will crash the US stock market this year. They're dumping SOTA open-source models on the US, nearly as good as the frontier labs, basically free. Nobody can refuse cutting a $100M AI bill down to $5M. Once that deal is on the table for everyone, Anthropic and OpenAI can't justify their prices, or their valuations. The bubble isn't AI. It's the price of AI. And it's about to pop.

@AndrewCurran_ This had to happen, and there are only two players: the USA v China. I think you might be wrong and it goes the other way: KYC for using certain models goes permanent.

I think both Fable 5 and GPT-5.6 get approved for general release next week, and for use outside of the United States as well. But people should remember this moment and remember this feeling, because it is almost inevitable that we eventually reach a point where approval does not arrive. Capitalism is going to tip the scales this time. I doubt they will approve one model and not the other, because doing so would be seen as incredibly anti-competitive. Fable and GPT-5.6 will probably receive the same clearance, probably on the same day. I also doubt they want to restrict sales outside the US, because that would be seen as anti-business and would trigger a major backlash against American closed-source AI. The rumblings of which you can already hear today. There is also a plan now taking shape on both the US left and right to create some version of an AI public wealth fund that pays a dividend directly to American citizens. That fund needs to be fed by the global sale of the big labs top models to people outside the US. So I think there will be no freeze on their use outside the United States this time. The other reason is that allowing this will make people happy, and it will soften the fact that Mythos, as was announced yesterday, is available only to a vetted group of US agencies and companies. I do not think that this basic structure will change from here on out. Mythos may eventually be made available to certain allies, but only after the US government, its agencies, and then some chosen American companies have access to Mythos-2, Sol-2, or whatever the new uber-model turns out to be. I do not think this gap ever closes again, not even for allies. And that means the US will increasingly possess an intelligence advantage that touches almost everything: voting, markets, corporations, academia, infrastructure, and the internal operations of foreign states. Having Mythos-n will always be trumped by whoever has Mythos-n+1. Anthropic themselves have said within nine months Mythos will look like a toy. That advantage, standing at the top of this tower, is too large to give up voluntarily. It also means that many things will become suspect. People will see shadows everywhere. Barring espionage, a deliberate leak, or the emergence of a non-US competitor at the top end of the scale, this structure will persist for some time. The public fight is about access to models. But the real fight is about access to the future. And from this point forward, whoever holds this power will also become increasingly capable of keeping it for themselves.

@markankcorn @exolabs And without network

@exolabs This is exactly what Nvidia announced last month with Rubin CPX — separating pre-fill from inference, using one chip with beefy compute and lower memory bandwidth then passing to another with far bandwidth and massive memory but slower compute

Clustering NVIDIA DGX Spark + M3 Ultra Mac Studio for 4x faster LLM inference. DGX Spark: 128GB @ 273GB/s, 100 TFLOPS (fp16), $3,999 M3 Ultra: 256GB @ 819GB/s, 26 TFLOPS (fp16), $5,599 The DGX Spark has 3x less memory bandwidth than the M3 Ultra but 4x more FLOPS. By running compute-bound prefill on the DGX Spark, memory-bound decode on the M3 Ultra, and streaming the KV cache over 10GbE, we are able to get the best of both hardware with massive speedups. Short explanation in this thread & link to full blog post below.