altryne1

Thanks Swyx! I agree, I mentioned in the end where I sit on the continuum, but I don't maintain any hugely critical production code.

Do you think the intended target will matter as much as it does in a year?

I spent the whole last week covering, speaking to, debating with many of the top AI Engineers, at the AIE Europe conference in London. One question came up, and I dubbed it the Z/L Continuum: Should AI Engineers even look at code anymore?

I love the aesthetics and the new approach here! Do you guys randomize which model appears in which panel to reduce left/right favoritism in developers? (silly Q I know but i was wondering)

This is so dope Chris! I wonder, can I continue sessions from my local Claude Code to Codespaces?

Will the rate limits go higher? How about other models? Qwen 2.5 is nice but 3 is nicer

MCP for every website? Global "Open" index? this is interesting

Everyone's saying 2025 is the year of agents, and keep saying this, but what does it mean?

Thanks Swyx for outlining what it means, where things realistically are and what are companies actually doing with agents!

I've used this + Groq yesterday to augment (with a chrome extension) the infinite fun game from Neal Agrawal, but generate actual images and not only emojis.

This feels like the future, near real time image and LLM generation (using Mixtral from Groq as my prompt writer) and Fal API for read time generation!

https://x.com/altryne/status/1760561501096575401?s=20

This is a special Sunday episode of ThursdAI podcast, which features an interview with GaragebAInd, the folks behind Platypus 70B-instruct, which is at present the best performing OS LLM finetune, and features AlignmnetLab which joined them and together they also have the best 13B LLaMa finetune called Orca Platypus 13B, which is very close to base LlaMa2 70B in performance.

Thanks for asking me to hop on a space together @swyx

I didn't imagine so many folks will tune-in, share their use-cases of these new superpowers, share limitations, explore the boundaries (no more running binaries) and then explore the jailbreaks (apparently binaries are still possible to run w/ some tricks!) and more!

Incredibly deep recap as well

Look at home assistant, it will come this year from them if anyone

I called the GEN-Z phone like and it pretty much blew me away in response speed. It replied often faster than my family from the other side of the world would!

I've been a reflect user since the beta, paid and then stopped.

Switched to mem (never paid them) and really liked the quickness with which I can just "dump" my brain into mem. But their are not customer focused, and seeing that they kept piling on VC money, felt wrong to me.

Then a friend suggested that I try Reflect again, I think after they added tasks, and I resubscribed and couldn't be happier. But I was, when they added whisper voice notes, and then implemented AI features, and then upgraded to GPT-4!

The customer attention is top notch, the interface was recently re-designed by the designer of tailwind UI, James.

It's a joy to use, super fast, stupid simple.

Previous user of both, but have opted to pay for Reflect, after Mem added AI features before adding a dark mode.

Any notes app that doesn't give a dark mode to their customers as high priority, no matter how cool, is not listening to their customers.

Alex has been obsessed with customer feedback, answering support emails, and on discord, and it shows. Reflect is beautiful, fast and super secure.

Mem has the twitter bot, which is really cool!

This is using auto_subtitle from https://github.com/m1guelpf/auto-subtitle with the medium whisper model

I'm one of the maintainers (in charge of the UI) for hlky webui repo! And we just updated our own colab, you can find it here : https://github.com/altryne/sd-webui-colab

GPT-4 is going to be released soon and blow everyone's minds again. And maybe you've been talking with GPT-3 in comments all this time /s

This is massive, not sure how this flies under the radar in HN