Recent content by ES Dev Team

Yeah i know a lot about open source models, been running them for over a year now, just not at this scale. GPT OSS 120b is a good model for programming, but is based on ChatGPT 4 era technology, so it is prone to hallucination. This means you really need to direct it. We run Step 3.5 Flash...

We just completed an integration of TipTap 3.0 into one of our software products. It's not something you get done in a day ( and also get right ) with AI. This was one of the hardest WYSIWYG integrations we've done. If i was to give these WYSIWYG editors a grade in terms of smoothness to use +...

Apple hardware: reasonably efficient gobs of ram ok speed ( m5 and up ) expensive little benefit in parallelizing multiple units Nvidia hardware: reasonably efficient but out of the box usually tuned to use 33% more power for 10-15% more perf ( if this is tuned down, it's very close to...

Personally.. i could care less about silicon valley. Their software quality going to crap lately and so much of the output is VC funded too, meaning it's guaranteed that they'll do a first one's cheap, last one's super expensive kind of pricing play. I got so tired of using their crap as an IT...

Nice to see the AI part working!

Our rule: ..and yes, we enforce it

I've seen 10,000's-100,000's of unique IP addresses at times, yeah, that wouldn't work for me. I'm not surprised that residential proxies walk through it!

Your word per token is off, a token can be as small as 1 character, but usually closer to 3 characters. a punctuation mark or space is usually a token too. We're still doing napkin math but 3x that figure But you may also be leaving out reasoning tokens, overhead of their system prompt, overhead...

To be clear: I think AI could do a lot of interesting things on a forum.. ..but the costs are nuts right now.. ..we really need the hardware to improve 2-3x in efficiency/cost. ( i say this as i sit next to a 50lbs AI box that can draw 900w from the wall and make the office into a sauna if a...

That's good. But let's think seriously about the cost of AI features. Per function that AI performs ( versus just using regular programming ), the resource cost is >1000x more The more you use it, the more your operations cost increases. At my dev shop, we found out how much hardware we need to...

I've heard of someone here having success with it

It's bad out there. On my magento site that gets attacked often now, i'm seeing a flood of residential proxies from USA cable providers. I unfortunately have to keep 'under attack mode' on :cry: Everyone is still blissfully running fail2ban. The only good alternative is to design a friendlier...

Honestly, amazing ratio of fradulent traffic to not. I Think this is not because cloudflare is great, but whatever forum this is is great :)

FYI i'm on the pro plan and i'm getting effective protection after some tuning. If they shove me into the business tier ( $200/mo ), i'm 100% building the alternative. So far my 31 other servers don't need cloudflare, they are not running ultra slow magento :)