nharada

Is there something specifically with Kimi that's better here? As far as I know Kimi pricing is about the same as Sonnet 5 -- what happens if you use that model and Fable instead? Or Grok 4.5 which is even cheaper?

A 24 kWh cap per day seems very reasonable. Drawing 8 kW is quite a lot.

why?

Is that the correct comparison? Fable is twice the price

Yeah if it has the amount of content people are expecting I'd say it'll be well worth 80 bucks

Imagine if Github said "if we detect you're building a competitor to Github, we will silently degrade the results of your CI actions so that tests sometimes randomly fail"

VB6 was the peak of desktop GUI app development change my mind

I've been wondering when we'll see it unambiguously showing up in there. I suspect this time next year it'll be visible for sure, maybe Q4 of this year?

Is the "etc" here "because of human greenhouse emissions, the earth is rapidly warming"?

At least here in SF the ideal thing would be that any vehicle dropping off in the bike lane gets fined or ticketed. This includes Waymo, Uber, cabs, personal cars, whatever. In practice it's very rare to get a ticket for this, which is why customers expect it from both Waymo and Uber.

In an era of 20% layoffs and entire orgs getting culled, this seems very reasonable

This is super cool and exactly what I've been looking for for personal projects I think. I wanna try it out, but the "agent" part could be more seamless. How does my coding agent know how to work this thing?

I'd suggest including a skill for this, or if there's already one linking to it on the blog!

Saying nothing about the actual performance of this model, it does strike me how .... minimal(?) this announcement is. Their safety section is like 2 paragraphs about bioweapons. Go look at the reports for OpenAI and Anthropic's model releases. It's like 50+ pages of tests, examples, reports, and benchmarks across a bunch of safety and wellfare metrics.

If Meta wants to be seen as a cutting edge massive lab they need to come across as one instead of looking like a school project version of a frontier model.

I wonder if this will actually be why the models move to "neuralese" or whatever non-language latent representation people work out. Interpretability disappears but efficiency potentially goes way up. Even without a performance increase that would be pretty huge.

I just can't find myself summoning the energy to be mad about markdown. It's good enough for like 99% of the things I use it for. Sometimes I get annoyed at specific extension support or whatever when I realize I shouldn't be using markdown for that task.

The devices are either dangerous, or they're not

That's not actually how it works though, it's all a risk and percentages. Nobody says "driving is either safe or it's not" or "delivering a baby is either safe or it's not"

Was expecting to be annoyed but this seems reasonable. You can have 2 power banks and can't charge them during flight

I’m curious if you think viewpoints have also gotten more extreme in this period. It feels like the gap in political ideologies has widened a lot since I was younger.

Surely this task must now be in the training data

Yeah as long as the chatbot is empowered to fix a bunch of basic problems I'm okay with them as the first line of support. The way support is setup nowadays humans are basically forced to be robots anyway, given a set of canned responses for each scenario and almost no latitude of their own. At least the robot responds instantly.

Nice, I like the idea. It sounds like qualitatively you haven't had any performance regressions while doing this, but have you tested it at all on any sort of benchmark or similar eval? I'm curious how well the actual system performs with less context like this. I mean it's possible it actually improves...

Another interesting thing here is that the gap between "burned out but just producing subpar work" and "so crispy I literally cannot work" is even wider with AI. The bar for just firing off prompts is low, but the mental effort required to know the right prompts to ask and then validate is much higher so you just skip that part. You can work for months doing terrible work and then eventually the entire codebase collapses.

+1

First, I agree with most commentators that they should just offer 3 modes of visibility: "default", "high", "verbose" or whatever

But I'm with you that this mode of working where you watch the agent work in real-time seems like it will be outdated soon. Even if we're not quite there, we've all seen how quickly these models improve. Last year I was saying Cursor was better because it allowed me to better understand every single change. I'm not really saying that anymore.

This is awesome and I'm really happy to see this progress. Landing a new chemistry in a production car THIS YEAR is some crazy velocity, especially compared to where other Na-Ion batteries are in the development cycle elsewhere. Is anyone else even close to having a car on the road with their cells?

The reason this is so exciting for me personally is for stationary energy. Because the raw materials are so abundant and have good cold weather performance, both grid and home level energy storage costs should come down significantly as this is commercialized further.

That's a massive jump, I'm curious if there's a materially different feeling in how it works or if we're starting to reach the point of benchmark saturation. If the benchmark is good then 10 points should be a big improvement in capability...

worry that the US will fall behind the curve

Man it's already over. It's hard to imagine the US autos EVER catching up at this point, even with state support.

I'm obviously biased, and I probably have more gripes than most about Waymo as a corporate entity, but the premise this article seems to be based on is "Waymo is a zombie company who will never release a real product" or something similar?

They seem to be scaling just fine. Here in SF they're ubiquitous and most people I know use them regularly (and usually prefer them to rideshare). Sure, it's not the type of growth possible with pure software, but they're doing 500k rides/week and are looking to be doing 1MM/week by the end of the year. What scale does this business need to be for the author to consider them a real company?

I think the assumption is valid. Most of the reasoning components of the next gen (and some current gen) robotics will use VLMs to some extent. Deciding if a temporary construction sign is valid seems to fall under this use case.

In that case working at a startup would be a thing someone would only do as a last resort, and the talent pool would consequently be extremely low quality. Sounds damaging to the scene to me.

It definitely feels like a jump in capability. I've found that the long term quality of the codebase doesn't take nosedive nearly as quickly as earlier agentic models. If anything it's about steady or maybe even increasing if you prompt it correctly and ask for "cleanup PRs"