aliljet

5 ms·

31.
by aliljet 6mo ago
How can you reasonably try to get near frontier (even at all tps) on hardware you own? Maybe under 5k in cost?
32.
by aliljet 6mo ago
Mythos is only real when it's actually available. If you're using Opus 4.7 right now, you know how incredibly nerfed the Opus autonomy is in service of perceived safety. I'm not so confident this will be as great as Anthropic
33.
by aliljet 6mo ago
I've found myself so deeply embedded in the Claude Max subscription that I'm worried about potentially makign a switch. How are people making sure they stay nimble enough not to get trarpped by one company's ecosystem over an
34.
by aliljet 6mo ago
How many levels of agents are here. Agents riding code by agents in a system driven by agents vibed by one lonely engineer in Redmond?
35.
by aliljet 6mo ago
The real problem is that scientists doing this sort of early work more often than not want to burn hardware under their desks. Renting infrastructure in Google cloud isn't the only way...
36.
by aliljet 6mo ago
I am hopeful that OpenAI will potentially offer clarity on their loss-leading subscription model. I'd prefer to know the real cost of a token from OpenAI as opposed to praying the venture-funded tokens will always be this cheap.
37.
by aliljet 6mo ago
I've been looking for broken 3090s for a short while. And the whole market is funny. Most of these devices have had their VRAM and GPUs physically harvested (for no clear reason). The ones that are truly broken, however, are still trad
38.
by aliljet 6mo ago
This is the rugpull that is starting to push me to reconsider my use of Claude subscriptions. The "free ride" part of this being funded as a loss leader is coming to a close. While we break away from Claude, my hope is that I can
39.
by aliljet 6mo ago
First, making sure to offer an upvote here. I happen to be VERY enthusiastic about local models, but I've found them to be incredibly hard to host, incredibly hard to harness, and, despite everything, remarkably powerful if you are wil
40.
by aliljet 6mo ago
What open models are truly competing with both Claude Code and Opus 4.7 (xhigh) at this stage?
41.
by aliljet 6mo ago
This is the reality I'm seeing too. Does this mean that the subscriptions (5x, 10x, 20x) are essentially reduced in token-count by 20-30%?
42.
by aliljet 6mo ago
I'm really curious about what competes with Claude Code to drive a local LLM like Qwen 3.6?
43.
by aliljet 6mo ago
wait. that's insanity. where did you get those numbers from? the 5x plan is obviously the right place to be...
44.
by aliljet 6mo ago
Have they effectively communicated what a 20x or 10x Claude subscription actually means? And with Claude 4.7 increasing usage by 1.35x does that mean a 20x plan is now really a 13x plan (no token increase on the subscription) or a 27x plan
45.
by aliljet 6mo ago
Have they effectively communicated what a 20x or 10x Claude subscription actually means? And with Claude 4.7 increasing usage by 1.35x does that mean a 20x plan is now really a 13x plan (no token increase on the subscription) or a 27x plan
46.
by aliljet 6mo ago
I'm broadly curious how people are using these local models. Literally, how are they attaching harnesses to this and finding more value than just renting tokens from Anthropic of OpenAI?
47.
by aliljet 6mo ago
I'm honestly curious who on these plans that's not working with an unlimited enterprise budget would even choose to burn real dollars like this beyond the subscription? What is the personal use case? It seems exorbitantly expensiv
48.
by aliljet 6mo ago
There's a weird 'token anxiety' you get on these platforms. And you basically don't know how much of this 'limit' you may consume at any time. And you actually don't even know what the 'limit' is
49.
by aliljet 7mo ago
I wonder how crazy the scale here can get. How far can I go? The bps.space guy is heading into space. Can the community hit the moon? Literally.
50.
by aliljet 7mo ago
As crazy as this seems, it's unlocking another variation of software engineering I didn't think was accessible. Previously, super entrenched and wicked expensive systems that might have taken years of engineering effort, appear t
51.
by aliljet 7mo ago
Are there evals showing how this improves outputs?
52.
by aliljet 7mo ago
There's a vibe in at least the PNW that feels like the tech sector is sloughing jobs and avoiding creating new ones courtesy of AI. I genuinely wonder if that feeling is backed by reality and whether it's large enough to be transl
53.
by aliljet 7mo ago
Does fine tuning really improve anything above just pure RAG approaches for usee cases that involve tons of direct document context?
54.
by aliljet 7mo ago
This is really what I want to understand. Where can we see real world performance benchmarks?
55.
by aliljet 7mo ago
Is this actually true? I want to see actual evals that match this up with Sonnet 4.5.
56.
by aliljet 7mo ago
How do you even pull away from a Gmail address? I'm nearly twenty years into that service. Getting banned would be absolutely devastating...
57.
by aliljet 7mo ago
I really really want to see how these images are starting to form into videos. The stills are clearly getting better and better, but what about when you need the stills to organically conform to a keyed script?
58.
by aliljet 7mo ago
The easiest way? Angelist.
59.
by aliljet 7mo ago
The path to declaring yourself accredited is uniquely easy. Just say it. The whole space is deeply unregulated and unaudited. What makes it insane is that those middleman are making a small fortune exploiting this loophole protecting large
60.
by aliljet 7mo ago
The public can absolutely participate in this by way of syndication deals. Those syndicates are what's covering up the true extent of ownership and they're essentially charging for access with their fees. It's oddly shady, po

More ›