We have been working in this space for the past year. Based on our experience, I no longer trust any claims unless they are backed by benchmark results (yes, benchmarks are painful to run reliably and expensive).

It is possible to reduce token usage. It’s just much harder than the basic approach.

One quick win is to just avoid wasteful tokens, for example run all the QA tools like the unit tests in --quiet mode, which only prints warnings/failures.

Thanks to coding agents, there is no reason not to

For anyone interested in private space flight: Wild Wild Space documentary. Also a commentary about startups in general.

Separately, I've been wondering how uneven human progress is. I think we have come a long way in electronics (relays to vacuum tubes to transistor to … to an M5 chip) but in some other domains, like moving objects through space, we are basically mostly reliant on thrust generated by burning fuel (EV's are a genuine advance, but mostly due to battery tech).

Ion thrusters say “hi!”

I have been wondering: more solar means we are making Earth less reflective. Solar energy that might have been reflected back to space is now absorbed by the PV to make electricity. But that delta has to go somewhere? Where does it go in terms of Earth’s temperature. Same with more nuclear… we’d be pumping more energy into the system.

Don’t take this as an argument against these technologies. It’s a huge win to not be increasing the CO2 concentration in the atmosphere by burning fossil fuels.

Yes, this does happen. Solar farms have an albedo modification effect. The good news is that compared to fossil powered generation, the albedo modification is "paid back" in less than a year due to lower GHG emissions:

"Small reduction in land surface albedo due to solar panel expansion worldwide"

https://www.nature.com/articles/s43247-024-01619-w

If you download the supplementary information PDF, figure 19 shows a plot of GHG payback times by site. The most common time is about 1 month. The greatest time they found was about 9 months. These times go down as solar panel efficiencies continue to creep upward, because solar panels in 2026 don't have lower albedo than in 2024, but they do have better light-to-electricity conversion efficiency.

We're actually heating the earth in two ways... one is by increasing the greenhouse gases in the atmosphere. The other is simply by using energy; any use of energy means moving something to higher entropy and thus producing heat, even if that energy comes from 100% clean or renewable sources. However, nobody talks about this second way of heating the earth because it's about two orders of magnitude behind the first, i.e. for now pretty insignificant.

As for making the earth less reflective with PV panels... yes, you are right and most of that absorbed solar energy directly contributes to heating the earth because PV panels aren't anywhere near 100% efficient in converting it to electricity, but even the part that is converted to electricity will become heat again in the end, like when it powers AI data centers. (If PV was 100% efficient, then the lack of reflection wouldn't directly do any heating, the heating would all happen when the electricity is used.)

For now that isn't a big concern because the greenhouse gases our civilization is releasing into the atmosphere do so much more heating than all our use of energy, but if we do become 100% renewable and survive that particular danger, then by Jevons' paradox we'll soon be using 10x to 100x more "clean" energy and be heating the earth all the same again.

Compare anthropogenic heat release to daily solar gain and radiative losses.

Then note that a higher temperatures will increase the radiative losses.

100x heat release and it's still about 1% of solar gain.

A solar array makes the earth less reflective by a fixed amount, for as long as the array exists.

Every unit of fossil fuel we burn makes the earth permanently less reflective (or at least something on the order of 100 years). You could compare a 1GW solar array (for its lifetime) to a 1GW coal plant running for x days, though I don't have the value of x.

Current levels of solar panel coverage on Earth are approximately 0%. If we ran the entire Earth on Solar power we’d need ~ half a million square kilometres, around 0.1 percent of the Earth’s surface.

Looking forward to the series finale this Friday!

Luna is at a very compelling point on the price/performance curve.

I have found that sometimes a smaller model with max reasoning is actually more expensive than using the next tier model with a lower reasoning effort. It’s certainly faster.

Agreed. with "Ultra" (higher than Max), the luna performs really well for my non-metric-backed personal experience

In my experience, the thing that makes your better at something is actually _doing_ it, badly at first and then gradually improving by iteration.

But for writing, I usually see the advice that says you should read a lot instead of write a lot. What gives?

You shouldn't read a lot instead of writing a lot, you should be doing both, in the same way a musician should practice their instrument and listen to a lot of music. Or, for a more HN-appropriate context, imagine trying to contribute to a FOSS project without reading any of the existing code or knowing the style guides. Your contributions could be fine, but they'll never be good.

I looked at the OTel schema generated for recording a single numerical metric. It was like 12 or 13 meta fields in addition to the actual metric fields like timestamp, metric and value.

OTel looks like something designed by a committee of committees, funded by someone who is in the business of selling cloud storage/data warehousing services.

Who are OpenRouter’s competitors?

OpenRouter doesn't decide on the pricing.

They do decide on the 5% markup. But as far as I can tell, all other routers just match 5%. Not sure what they're competing on.

Yes but Sol dropping in price by 50% is not OpenRouter deciding. It's OpenAI.

I don’t think that’s true. OpenAI docs don’t have this price change. I assume they’d be the source for this post if it was true. The banner on OpenRouter for me says Gemini 3.7 discounted for a limited time, but if I click through that I get to this page: https://openrouter.ai/models?discount=true

That shows a bunch of models, including Sol, with a discount. None of them say how long it’s for, but I’d assume in all their cases it’s for a limited time as the banner said, and only on OpenRouter.

OpenRouter margins are not 50%.

As I understand OAI is offering discount only for users using the model via OR. Not sure why. Maybe they want OR users to try the model and switch to OAI subscription or something.

It might also be worth noting that the discount is only if OpenAI itself is used as the provider via OpenRouter. The discount does not exist for Azure or Amazon hosted Sol.

I posted elsewhere, but the Azure uptime & performance for Sol is truly dire. OpenAI is offering 5x faster latency, 4x faster tokens generation, and vastly better uptime (Azure US has only 87% uptime), all for 50% of the price now. I assume the pricing is to compete with other shiny new models (Grok, Qwen etc), but it might also be to cut-off a truly poorly performing Microsoft hosting experience.

Yes, but this discount must be coming from OpenAI and not Open Router.