371–380 of 600 posts
Earlier quoted context omitted.
Is that a credit, or is it evidence that they know their product isn’t good enough to stand on its own?
This is nothing surprising and not unique to OpenAI. Marketing is more than half the game for any product.
Earlier quoted context omitted.
I have an Asus GX10 that I run Qwen3.5 122B A10B on, and I use it for coding through the Pi coding agent (and my own); I have to put more work in to ensure that the model verifies what it does, but if you do so its quite capable. It makes using my Claude Pro sub actually feasible: write a plan with it, pick it up with my local model and implement it, now I'm not running out of tokens haha. Is it worth it from a unit…
Thank you. I've been using ollama for a much more modest local inference system. I'll research some of the things you've mentioned.
Earlier quoted context omitted.
Call it fall for it, but here are my two experiences, with both applications open. ($20/month plan for both) - Claude: Good for ~20 minutes of work once every 4 hours - Codex: Good for however long I want to use it. Claude nerfed their product so that it's not usable, so I use something else.
Since we’re sharing anecdata: I also have the $20 month plan for codex, and I hit the five hour limit after about an hour of work every single time I open it. I use it for personal side projects primarily in the evening after kids are in bed, so my strategy is to launch it about 4pm and send a simple prompt to prime the 5 hour window to end at 9pm, start working about 8pm, and then I can use up the existing 5 hour wi…
Just reading the comments here it's amazing how many people seemingly don't know that Claude Desktop and Cowork basically already does all of this. Codex isn't pioneering these features, it's mostly just catching up.
Earlier quoted context omitted.
What tasks do you use qwen3 for? Coding? Are you running it on CPU or GPU? What GPU does that Framework have? Thanks!
I have an Asus GX10 that I run Qwen3.5 122B A10B on, and I use it for coding through the Pi coding agent (and my own); I have to put more work in to ensure that the model verifies what it does, but if you do so its quite capable. It makes using my Claude Pro sub actually feasible: write a plan with it, pick it up with my local model and implement it, now I'm not running out of tokens haha. Is it worth it from a unit…
Earlier quoted context omitted.
Since we’re sharing anecdata: I also have the $20 month plan for codex, and I hit the five hour limit after about an hour of work every single time I open it. I use it for personal side projects primarily in the evening after kids are in bed, so my strategy is to launch it about 4pm and send a simple prompt to prime the 5 hour window to end at 9pm, start working about 8pm, and then I can use up the existing 5 hour wi…
What kind of side projects do you need to run these models for that many hours? I haven't experimented with Opus to that extent and mostly supervise it and/or am prompting it every 5-10min to fix something up.
- sysadmin tasks for my home server which runs home assistant, plex, and minecraft servers. Being able to tell it "Set up a minecraft fabric server with this list of mods" is pretty nice, and it's fairly competent at putting together home assistant dashboards and automations (make sure you have backups of anything it's allowed to touch, though--it may delete stuff without warning).
- Several small web apps primarily for my own use.
- Currently working on an opinionated desktop writing app for my own use.
There seems a fair enthusiasm in the UI of these to hide code from coders. Like the prompt interaction is the true source and the actual code is some sort of annoying intermediate runtime inconvenience to cover up. I get that productivity can be improved with a lot of this for non developers, just not sure using 'code' as the term is the right one or not.
The fact that the Codex app is still unavailable on Linux makes me think the target audience isn't people who understand code.
Just reading the comments here it's amazing how many people seemingly don't know that Claude Desktop and Cowork basically already does all of this. Codex isn't pioneering these features, it's mostly just catching up.
:^)