Footprint0521

I am currently stacking 5 opencode go accounts, spending $50 a month for $500 in deepseek v4.1 flash credits, using OMP as my harness

Or you could use the server console commands, for a time compressed equivalent effect

Is this something that you would use in coffee shop and exchange for an external computer monitor? I have been really on the fence about investing in something like this

Welcome AI2027 nightmares lol

Hey if you’re OpenAI, you’ll get free WiFi and no legal repercussions!

Cue the movie Mercy lol

I have mine in a RAID 5 setup rn because I was scared about that, but I definitely should set up offsite backups, it’s just so expensive

What happened?? As someone who is currently doing this without backups…

Dude do you have any screenshots of the iOS app or website?

How do you use omp with subagents and not burn through credits? I tried using orchestrate and it drained my wallet… I didn’t know about /tan though, that is sick

Ashai linux?

This is only for people who haven’t personally used deepseek v4.1 flash, or MiMo V2.6, and don’t know about cost per task on artificial analysis

Dude, hop on the Gemini student plan and try Gemini Notebook and Gemini Live. Live is a personal TA, Notebook literally podcasts the readings and what you need to learn

That’s what I have been using, but from glancing through this repository, it looks like the win here is that you can define where you want things and yet still have diagram as code

WALL·E

The moment AAA gaming becomes possible on macOS, the floodgates open and windows is dead

Three.js with cannon.js is what you want, but the more you work with it, the more you realize how the difficult part is not so much the “game” or “engine” in that sentence but the word “web”… optimizing assets is a massive skill all in its own to be storage friendly and usable without the easy default of laggy and massive download size

Just curious, what is your ram and how many tokens per second do you get with that?

Yeah that sucks… unless it’s over the top export controls for DoD work that really doesn’t make sense

Kimi code with k256 (I’ve heard you can easily one shot implement a proxy to other providers, even with deepseek v4 flash, if you don’t want kimi code) has stupidly low rate limits for and cost, compared to OpenAI which has massive rate limits and more cost

That is a wild article! Hopefully the recent US open source model initiatives can pan out to cut costs, because I’m not a fan of dependence on Chinese models either, but I’m in a spot where I need to maximize intelligence on minimal cost…

There’s not even a distant comparison US lab rn which is sad

Dude $3k? Holy heck you should look into K3/Deepseek V4 Flash

Facts lol, now all the Chinese models are 1/40th of the cost for the same intelligence

Genuine question, what made you pursue C++ with this and not Rust?

Mythos alone proves the future of cybersecurity and software is agentic, and if you don’t believe that, run it back in 2 years when you are fired for someone who does..

+1 at least he cited his sources lol

Nooo, now you can’t jailbreak it :,(

That’s what I did before I looked into KOReader and now I’m sad I did

Playing devils advocate, does from scratch really matter if all frontier labs are training off each other anyways? Practically speaking, businesses/consumers just care about lowest inference cost for maximizing intelligence anyways (not to mention Anthropic/OpenAI forcing KYC/litigation barrier trash for access to any cybersecurity/IT capabilities) that I literally just cancelled Claude today.

Yeah it’s sad the CCP has my information. But it’s either them or the feds, and at least Chinese models actually work for cybersecurity tasks, not to mention aren’t stupid expensive

Artificialanalysis.ai rn on opus 5 is a joke. The main intelligence benchmarks it is like 1% better than Fable, but the cost difference between that and K3 is so funny lol. It’s the same with cars— you don’t have to do it from scratch, as long as you can do it cheaper and with the same quality, hence Toyota/Honda taking over

Yeah it’s sad that American models are censored more than Chinese ones lol (outside of asking them about the CCP) but it’s where we are at I guess :(

Booz Allen is cute, but if china can train K3 on a fraction of the US compute availability, yet it benchmarks almost equivalent to Fable for a third of the cost, it’s game over for US labs in the long run.

Believe me, I wish this wasn’t the case, but open up the hardware on the device you are reading this on and tell me the majority of tech inside wasn’t made in China….

Manufacturing was lost a long time ago, this is really just another way

Real… I’ve been using Deepseek v4 pro max as my main and then k3 in web (more usage credits) for automating what my deepseek agents do

Switch to K3 and you won’t look back, I promise!! I got so fed up with Claude and finally bit the bullet to switch and it’s amazing