so apparently they have custom hardware that is basically absolutely gigantic chips - across the scale of a whole wafer at a time. Presumably they keep the entire model right on chip, in effectively L3 cache or whatever. So the memory bandwidth is absurdly fast, allowing very fast inference.
It's more expensive to get the same raw compute as a cluster of nvidia chips, but they don't have the same peak throughput.
As far as price as a coder, I am giving a month of the $50 plan a shot. I haven't figured out how to adapt my workflow yet to faster speeds (also learning and setting up opencode).
For $50/month, it's a non-starter. I hope they can find a way to use all this excess bandwidth to put out a $10 equivalent to Claude Code instead of a 1000 tok/s party trick I can't use properly.
GLM-4.6 is on par with Sonnet 4.5. Sometimes it is better, sometimes it is worse. Give it a shot. It's the only model that made me (almost) ditch Claude. The only problem is, Claude Code is still the best agentic program in town and search doesn't function without a proper subscription.
Cerebras offers pay-per-token. What are you asking for? Claude Code starts at $100, or $15/mtok. Cerebras is already much cheaper, but you want it to be even cheaper at $10?
Yes this is the output speed. Code just flashes onto the page, it's pretty impressive.
They've claimed repeatedly in their discord that they don't quantize models.
The speed of things does change how you interact with it I think. I had this new GLM model hooked up to opencode as the harness with their $50/mo subscription plan. It was seriously fast to answer questions, although there are still big pauses in workflow when the per-minute request cap is hit.
I got a meaningful refactor done, maybe a touch faster than I would have in claude code + sonnet? But my human interaction with it felt like the slow part.
I find AI is most useful at the ancillary extra stuff. Things that I'd never get to myself. Little scripts of course, but more like "it'd be nice to rename this entire feature / db table / tests to better match the words that the business has started to use to discuss it".
In the past, that much nitpicky detail just wouldn't have gotten done, my time would have been spent on actual features. But what I just described was a 30 minute background thing in claude code. Worked 95%, and needed just one reminder tweak to make it deployable.
The actual work I do is too deep in business knowledge to be AI coded directly, but I do use it to write tests to cover various edge cases, trace current usage of existing code, and so on. I also find AI code reviews really useful to catch 'dumb errors' - nil errors, type mismatches, style mismatch with existing code, and so on. It's in addition to human code reviews, but easy to run on every PR.
Wow, 30 minutes to rename functions and tests? I wonder how much energy and water that llm wasted for something that any lsp supporting editor can do in a second.
Settings > Extensions > Git
Blame: Status Bar Item Enabled (check this)
Blame: Status Bar Item Template (use this value)
${authorName} (${authorDate}) ${subject}
There's no secret sauce, all these variables are shown right above the input.
weird - I had never heard of Jellycats as a brand, but last night I went to go buy a second copy of my toddler's favorite lovey (since the first will inevitably get ruined at some point... thinking ahead). And it was a jellycat brand, must have been gifted to us.
* I like how the map moves around. It helps nail down relationships to neighbors
* I don't mind a few extra "Where's canada", even though it's not that useful
* I'd like the pause between answers be shorter.
* Small countries are impossible to see when zoomed out on the first exposure, even when selected right. I find myself knowing the area it's in (ie, central america) but not which exact country. So selecting it right when zoomed out doesn't get me the correct answer.
A dipole is a very simple antenna, and pretty efficient. There are other shorter options that have worse performance too, but perhaps more suitable for an adhoc tree deployment. (random wire, EFHW)
(I know you know this, but just adding in. Ham is fun! I like doing park & camping deployments)
In 2021 I lost a good chunk of weight the old fashioned way. From 250ish to 215. I did it with "good old fashioned working out and not eating more". It was a miserable, white-knuckle experience. I was eating healthy food, enough calories (moderate but sane deficit), but the only thing I thought about at all moments was getting to the next meal. What snack is low enough calories to have to make it. It was miserable. As soon as I let up a bit, everything unraveled and I found myself back in the 250s by the start of this year.
Now I'm on Tirzepatide (Zepbound), and I'm back to 235ish, and trending lower. I still work on eating healthy, but now I'm not just HUNGRY at all moments. My life continues, and I only have to make individual healthy choices at meal times, and grocery times, rather than a constant struggle at all waking moments. It's seriously a big difference.
> I still work on eating healthy, but now I'm not just HUNGRY at all moments.
I think this is something a lot of people pushing back against the GLP-1 agonists don't realize because they don't experience it: back before I started Mounjaro (another GLP-1 agonist) I was constantly hungry if I hadn't eaten a meal in the last 45 minutes. Absolutely zero hyperbole there - I once went to an all you can eat buffet, ate until I was over full, came home, and within about an hour and a half of that I was snacking on something because I was hungry. Not peckish. Not "feeling like a snack". Hungry to the point where that feeling intruded on my every thought until it was sated.
After starting Mounjaro that's GONE. Gone gone. I now have to set an alarm to remember to eat. It's absolutely phenomenal and likely the reason why I'll live past my forties instead of being stuck in that same cycle and dying of the effects of obesity.
> Absolutely zero hyperbole there - I once went to an all you can eat buffet, ate until I was over full, came home, and within about an hour and a half of that I was snacking on something because I was hungry.
I don't have any eating issues but that reminds me of the first time I went on a 7-day cruise.
There's nothing to do on the ship, and the food is free and pretty tasty, so... I basically ended up at the buffet eating and drinking all day long. Sausage and egg biscuits, banana bread, pot roast, steak, pasta, fried rice, cinnamon buns, they had everything. I was stomach-busting full, every minute of every day. I'd gorge myself on a huge plate of Indian food from the buffet, and then a few hours later head to another deck for a lobster dinner. Not to mention, drinking coffee, beer, and wine the entire time.
It was kind of insane. And what was crazier was after a few days of this routine I got used to it, and even looked forward to eating more food the next day. It was sort of like directly embracing one of the seven deadly sins to the maximum extent possible. I'm not sure what that experience means other than it seems like the the human body can comfortably arrange itself into a habitual downward cycle fairly rapidly.
It's because we're evolved for boom-bust cycles. Give it another 500,000 or so years and humans might evolve to cope with food always being available at all times.
Not only that, but type 2 diabetes makes you paradoxically more hungry - your body thinks it is starving because it cannot get sugar into its cells so it makes you MORE hungry which causes you to get heavier which often causes the diabetes to get worse which makes you MORE hungry which means you eat and get heavier and......
As someone who's never struggled with weight, it's been eye opening to hear how food focused a lot of peoples thoughts are. It was like on the same level as finding out some people can't visualize things in their minds.
It's more expensive to get the same raw compute as a cluster of nvidia chips, but they don't have the same peak throughput.
As far as price as a coder, I am giving a month of the $50 plan a shot. I haven't figured out how to adapt my workflow yet to faster speeds (also learning and setting up opencode).