rapind

Rechecked my benchmarks. 193,000 writes per second (rust), each write holds the lock for about 5 µs.

Full report here: https://github.com/pairshaped/gleam_sqlite_benchmarks/blob/m...

Obviously real world queries will vary, but I tried to be representative of the type of queries / updates I am sending.

It should be yes. Well it's ruby, so you could probably overwrite it with your own `def paid?`, but I suspect the `&& !refunded?` is just cruft.

I'm using it w/ a Rust server and I can get up to 80k writes per second, 200k reads per second. You won't get that with other databases without paying a lot and doing a lot of special architecture.

The problem is "remote" SQLite. Just run a "single" beefier box with in-process SQLite and Litestream backups to object storage and never have to think about RDBMS again. I'd wager there are 90%+ applications that would be better served, more performant, and cheaper with this design.

I'm able to handle around 200k/s on a sub $100 box. https://github.com/pairshaped/gleam_sqlite_benchmarks/blob/m...

Crazy that we need to keep relearning this over and over again.

Yes it's a problem. Not dismissing it. Just a less expensive problem than CPU or RAM is all I'm saying. It's been a thorn for me too (being on a macbook myself, but now I have a dedicated build box).

RAM prices are insane right now...

Yeah, there are some things I've done to mitigate this, especially for the dev loop. Moving templates out of rust (losing some safety). Splitting the application into vertical slices. Only generating code that's changed. Turning features into crates.

It's been a journey...

Both (and memory) are a problem. Build sizes is definitely more easily solved just by throwing more disk at it though (unless you are using a hard constrained mac...).

I would assume hardware production has been ramping up significantly and it's more of a temporary supply chain issue though. While it could last multiple years, that's certainly not long enough for the hyperscalers to get a return on investment. Obviously more complicated than this, but still... I'm not buying.

We're also starting to see efficiency gains, so we might meet halfway.

> There are canadian inference providers running deepseek

As someone who's in the market for Canadian inference (data sovereignty reasons), the pickings are slim for DSF 4.1 man.

Spoiler, it's not clear by any means. There's no evidence to back up the claim, it's just their feelings... which is fine, but I definitely wouldn't call it clear.

KegWorks is an option.

Even so, scale, severity, and flaunting of the law matter. Current political powers have turned it up to 11. You can absolutely "both sides bad" in general, but I don't think you should be doing so within this context or it's disingenuous / whataboutism.

Is Russia still under US sanctions though? I read something about a Trump deal to buy diesel from Russia.

It's entirely possible that candidates who are not promising real consequences will be unelectable in the next democratic primaries.

> It depends on how you use it.

I suspect you are right. For context, I was assuming a 20x w/ OpenAI or Anthropic subscription as the comparison (or both). dsf 4.1 was going to run me about 2-3 times the cost of either of those for the same amount of work. Obviously you can optimize differently, but that's true of subscriptions too. I was using pi and had it evaluate it's ideal context compaction point based on usage and API rates. Keep in mind though I was using a ZDR provider, so slightly higher costs. If I want them to train on my code, I could shave a few $ off.

That's not even taking into consideration all of the resets you get from the frontier subs. Which lately seems to at least double usage (more like 5x recently with OpenAI if you count the credit grants). But OpenAI is tweaking it's pricing, so it's always a moving target... which is kind of annoying until you learn to just ignore it.

It's really not cheaper than frontier subscriptions. It's getting closer, and it's a great model, but it is not more value per task than the frontier subscriptions. Don't be swayed by the token costs, it's very chatty, like 3x more tokens for the same task as sol. I used dsf 4.1 full time for about a week.

It blows frontier API pricing out of the water, but again, look at cost per task, not token usage. Still easily wins though for my work.

I do think it's the most viable alternative I've seen so far, and that applies pressure to the frontier models. Should subscription prices hike or become unavailable for some reason, I know what I'll be using.

When pricing this, it's important to consider whether or not you want to opt out of data training. You won't get the advertised rate. Also the dsf 4.1 subscription providers are throttled af... and of course they are, because otherwise they'd be haemorrhaging money.

Based on the comments in this thread, you'd think that Reddit was down.

Exactly. It's just like when we used to pay workers a living wage. Well where did all of those jobs go amiright?! Stupid peasants can be so annoying... we're just trying to bring everyone into the grand age of abundance!

Totally not a new robber baron era at all. That's just fake news guys.

/s

> One thing i don’t understand is how pro-surveillance advocates don’t understand that the moment they are not behind the terminal, they are the target.

I think it's three things; (a) They're just being paid not to care, (b) short-termism, and (c) are naively hoping they get to stay in the club on the other end of the surveillance. Hard to convince nearly anyone to give up short term profit for long term societal benefits.

Cynically I worry we are seeing the decline of democracy and our new overlords are seeing that too (many accelerating it). Surveillance is necessary for control. The AI, mass surveillance, and drone combination is getting pretty scary TBH.

It’s honestly my biggest issue right now during development. I’m refactoring into a “vertical slice” architecture and then will try crating by feature… fingers crossed.

The Meta contributor tier is explicitly very cheap because they keep and train on your data. If you opt out of data collection and training then you don’t get the contributor rate and it’s bad value.

I think they’re saying they would rather use Muse or Dots.

I am experimenting with this in my project right now. I haven't reached a verdict yet, but in theory it sounds good. I think you're still paying the linking cost though?

I agree, and I started sending well defined review packages to Luna x-high instead, which is how I have this setup when using Codex (Sol drives, Luna async reviews, and Sol keeps moving). Same process with DS really dropped my DS usage by a lot and I think that might actually be the secret sauce. Especially if you use Luna through a subscription (probably the entry pro level would be enough). I'm not sure what a Luna equivalent open model is though. 5.6 Luna max was catching a lot of issues while my implementer keeps rolling. Last couple of days I had 3-4 Sol Mediums running with Luna x-high reviewers (async reviews) and one Sol medium orchestrator and Astra X-high to plan everything out.

Gotta be honest though, I don't love fiddling with this all the time. I would rather be working on my projects than evaluating my usage. Having DS in my back pocket should i need it is a relief. The providers get fiddly though too.

That's an unfortunate experience. Think of v4.1 flash as actually v5.0 flash. It's night and day compared to the 4.0 flash (and 4.0 flash was unintuitively better than 4.0 pro). I would re-evaluate with v4.1 flash. I'm not saying it better than Sol or anything, but it's in the ballpark.

Fireworks directly. At the time they were the best value of cost, speed, ZDR. They got slower on me though, but I think they are retooling, so maybe things have or will get better again. I think fireworks is primarily for when you want to do your own training on top, which I wasn't doing.

I found omp used quite a bit more tokens than my fairly basic pi setup... but most of those tokens would be cached with DS V4.1, so maybe worth if there are gains elsewhere.

pi. I wasn't even going that hard. I checked the logs for Sep 18 and I did just shy of 3b input with approx 98.5% cache and 5.8m output, which cost around $35. Most of the was a Rust code review exercise with 1 driving agent and a varying number of subagents (up to 6 some times). I do the same with with Sol med/high driving and Luna x-high reviewing and get at least as much done if not more in a day, but I'd use up two x20 weekly allowances for the week. Worth noting that token cost isn't super meaningfull on it's own, because DS is super token heavy (but also great at caching) compared to Sol. (my stats show DS uses 3x the tokens as Sol)

The shape of my work changes obviously, so it'll vary, sometimes more, sometimes less. For example, fixing all of the bugs and defects I found that week was 2-3 times the effort and chewed through my ChatGPT allowance, but I had banked resets...

Also worth noting that codex models have been kind of all over the place recently with their usage... and it looks like costs are changing again.

So I took Deepseek V4.1 Flash for a spin maybe 2 weeks ago now (before Luna 6 and Sol 6 were announced), and I racked up $100+ in about 2-3 days. It was pretty great, but it uses way more tokens (TPS is fast, but it's way more tokens per turn) than Sol 5.6 which I found to be about it's equivalent at the time (on medium or high, with DS on max). My cache rate was around 98-99%.

It would definitely cost me more per month than a x20 ChatGPT or Claude plan, probably around $400+ was my estimate at the time. This was with Fireworks (ZDR) which has since increased their prices (and got slower!).

That being said, very impressed with the model, and looking forward to what comes next. As the frontier models become less subsidized, the open models will become more appealing.

P.S. There are subscription plans for open models, but I've found most of them to be extremely slow, have model throttling (only so much of model X), and also very sketchy about training and data retention. No thanks! If you want to share your data, just use Muse Spark contributor. Seems impossible to beat that on price per task if you don't mind feeding your data to the Meta machine (spoiler: I won't).