Prof Adebayo

4.9K posts

Prof Adebayo banner
Prof Adebayo

Prof Adebayo

@ProfAdebay

Founder @Academicnight, Founder @InstitutionApp | i literally want ai to replace you.

you can't reject light forever Joined Mayıs 2023
902 Following1.2K Followers
Prof Adebayo
Prof Adebayo@ProfAdebay·
Just 2 prompts on Opus 4.6 using Claude Code… My 4th Codex subscription is about to hit its weekly limit, so I switched to Opus for iteration. It seem I have to sub for 5th codex account 🙆
Prof Adebayo tweet media
English
0
0
2
83
Jen Zhu
Jen Zhu@jenzhuscott·
We should all take a moment to appreciate @karpathy - he could’ve built any of his experiments/realisations into a company, raise billions, & milk subscription fees from all of us; then hype the myth to create an illusion his walled gardens r impossible to reach by us mere mortals. Instead, he just dumping his knowledge and knowledge-how and share with everyone openly. No myth, no hype, no profiteering. A true giant and leader ♥️🔥♥️
Andrej Karpathy@karpathy

Wow, this tweet went very viral! I wanted share a possibly slightly improved version of the tweet in an "idea file". The idea of the idea file is that in this era of LLM agents, there is less of a point/need of sharing the specific code/app, you just share the idea, then the other person's agent customizes & builds it for your specific needs. So here's the idea in a gist format: gist.github.com/karpathy/442a6… You can give this to your agent and it can build you your own LLM wiki and guide you on how to use it etc. It's intentionally kept a little bit abstract/vague because there are so many directions to take this in. And ofc, people can adjust the idea or contribute their own in the Discussion which is cool.

English
35
36
664
65.8K
Prof Adebayo
Prof Adebayo@ProfAdebay·
Calm down…😂 He actually has a solid project. If not for the Twitter bubble, it would already be recognized as valuable here. The reality is, it is valuable, just not to devs (Twitter devs). Outside Twitter, especially among people who understand don't know SMS gateways, it is gold fr them & SMS gateways is open source for a reason: they’re meant to be forked, extended, and built into something even better.
English
1
0
2
29
kedimuzafer
kedimuzafer@kedimuzafer·
@iamZenderock @heynavtoor Sms gateway is an open source project. What's more modern than this? I've been testing with sms gateway and it's pretty successful and free
English
2
0
1
337
Prof Adebayo retweeted
Nav Toor
Nav Toor@heynavtoor·
🚨 Twilio charges $0.0079 per SMS. Someone just turned any old Android phone into a free SMS gateway. Unlimited messages. $0. It's called SMS Gateway for Android. Install it on any Android phone. It becomes a full SMS sending and receiving server with an API. No Twilio. No MessageBird. No per-message pricing. No contracts. Just an old phone and a SIM card. Here's what's inside this thing: → Send and receive SMS through a REST API from any app or service → Works with any Android phone running 5.0 or newer → End-to-end encryption. Messages are encrypted before they leave the device. → Multi-SIM support. Use multiple SIM cards on one phone. → Multi-device support. Connect multiple phones to the same account. → Real-time webhooks for incoming messages → Multipart messages with auto-splitting for long texts → Track delivery status of every message in real time → No registration required. No email. No account in local mode. Here's the wildest part: That old Android phone in your drawer that you haven't touched in 2 years? Install this app. Insert a SIM card. You now have your own private SMS infrastructure. Two-factor authentication. Order confirmations. Appointment reminders. Notification alerts. All the things startups pay Twilio thousands a month for. Free. Running on a phone you already own. Startups spend $500 to $5,000/month on SMS APIs. This costs the price of a SIM card. 875 GitHub stars. 359 commits. Apache 2.0 License. 100% Open Source.
Nav Toor tweet media
English
287
964
9K
801.4K
Prof Adebayo
Prof Adebayo@ProfAdebay·
@theo Probably the Anthropic ban has reached him too.
English
0
0
0
350
Prof Adebayo
Prof Adebayo@ProfAdebay·
@chrisparkX @karpathy .. except you’re gating it against unauthorized business use. But X is nw somewhat like Google, so not everything should be gated, or at least some parts should be relax.
English
0
0
0
28
Prof Adebayo
Prof Adebayo@ProfAdebay·
@chrisparkX @karpathy If “read” endpoints are enabled, I don’t think there’s any real risk, even if an agent spams them.. since they’re not performing any write operations. So there’s no need to supergate the read endpoints.
English
1
0
1
252
Chris Park
Chris Park@chrisparkX·
We’ve made major upgrades to X API: • Pay-Per-Use now GA worldwide • XMCP Server + xurl for agents • Official Python & TypeScript XDKs • API Playground - free realistic simulations New releases coming will be a game changer. Start building → docs.x.com 🚢
Elon Musk@elonmusk

Try using the X API

English
341
254
3.1K
66.2M
Naz Bent
Naz Bent@Bent302·
@ProfAdebay I pulled the server out of a scrap bin. The RAM, 3060, and the CPU were purchased Q1 2025. Everything else was this year.
English
1
0
1
35
Naz Bent
Naz Bent@Bent302·
I have no clue how this monstrosity even booted. RTX 3090 FE 24GB 2x RTX 5060TI 16GB (32GB) Quadro RTX 5000 16GB 72GB total VRAM 256GB system RAM 22 core 44 thread intel xeon processor All in cost ~$3000 #opensource #localllama #qwen #ai #nvidia
Naz Bent tweet mediaNaz Bent tweet mediaNaz Bent tweet media
English
8
0
36
14.5K
0xSero
0xSero@0xSero·
GPT-5.3-Codex is still the best coding agent, no doubt about it. GPT-5.4 is better at computer use, but doesn't match the sheer autistic power Codex holds.
0xSero tweet media
English
63
14
894
79.7K
Tj Dunham
Tj Dunham@RealTjDunham·
@DSBatten now what if that same device could run ai inference, contributing its compute towards running the biggest models trustlessly been building something like that, i believe its what satoshi wouldve wanted for ai models
English
2
0
3
192
Daniel Batten
Daniel Batten@DSBatten·
Bitcoin grew out of the Internet, but has now become more resilient than the Internet. A $6 feature phone, a 2G network, even mesh radio (like a low-tech walkie-talkie) can send money anywhere in the world. Fee-free. Bank-free. Middleman-free. Delay-free
Bitshala@bitshala_org

No internet? No problem. Bala just dropped the ultimate cypherpunk demo at the BOSS Summit: From Off-Grid to On-Chain He literally broadcasted a live Bitcoin transaction using Mesh Radio. No ISPs, no Wi-Fi, no cellular data. Just pure radio waves bypassing the traditional internet layer to hit the mempool. When we say we are building unconfiscatable money for uncertain times, this is exactly what we mean. Mind blown. Here is the repo used to connect meshtastic to bitcoin core:github.com/BTCtoolshed/Me…

English
19
122
720
43.2K
Prof Adebayo
Prof Adebayo@ProfAdebay·
Honestly, this would be really cool with local LLMs. In fact, it can be used as a memory layer for local dense LLMs.
Farza 🇵🇰🇺🇸@FarzaTV

This is Farzapedia. I had an LLM take 2,500 entries from my diary, Apple Notes, and some iMessage convos to create a personal Wikipedia for me. It made 400 detailed articles for my friends, my startups, research areas, and even my favorite animes and their impact on me complete with backlinks. But, this Wiki was not built for me! I built it for my agent! The structure of the wiki files and how it's all backlinked is very easily crawlable by any agent + makes it a truly useful knowledge base. I can spin up Claude Code on the wiki and starting at index.md (a catalog of all my articles) the agent does a really good job at drilling into the specific pages on my wiki it needs context on when I have a query. For example, when trying to cook up a new landing page I may ask: "I'm trying to design this landing page for a new idea I have. Please look into the images and films that inspired me recently and give me ideas for new copy and aesthetics". In my diary I kept track of everything from: learnings, people, inspo, interesting links, images. So the agent reads my wiki and pulls up my "Philosophy" articles from notes on a Studio Ghibli documentary, "Competitor" articles with YC companies whose landing pages I screenshotted, and pics of 1970s Beatles merch I saved years ago. And it delivers a great answer. I built a similar system to this a year ago with RAG but it was ass. A knowledge base that lets an agent find what it needs via a file system it actually understands just works better. The most magical thing now is as I add new things to my wiki (articles, images of inspo, meeting notes) the system will likely update 2-3 different articles where it feels that context belongs, or, just creates a new article. It's like this super genius librarian for your brain that's always filing stuff for your perfectly and also let's you easily query the knowledge for tasks useful to you (ex. design, product, writing, etc) and it never gets tired. I might spend next week productizing this, if that's of interest to you DM me + tell me your usecase!

English
0
0
0
21
Farza 🇵🇰🇺🇸
Farza 🇵🇰🇺🇸@FarzaTV·
This is Farzapedia. I had an LLM take 2,500 entries from my diary, Apple Notes, and some iMessage convos to create a personal Wikipedia for me. It made 400 detailed articles for my friends, my startups, research areas, and even my favorite animes and their impact on me complete with backlinks. But, this Wiki was not built for me! I built it for my agent! The structure of the wiki files and how it's all backlinked is very easily crawlable by any agent + makes it a truly useful knowledge base. I can spin up Claude Code on the wiki and starting at index.md (a catalog of all my articles) the agent does a really good job at drilling into the specific pages on my wiki it needs context on when I have a query. For example, when trying to cook up a new landing page I may ask: "I'm trying to design this landing page for a new idea I have. Please look into the images and films that inspired me recently and give me ideas for new copy and aesthetics". In my diary I kept track of everything from: learnings, people, inspo, interesting links, images. So the agent reads my wiki and pulls up my "Philosophy" articles from notes on a Studio Ghibli documentary, "Competitor" articles with YC companies whose landing pages I screenshotted, and pics of 1970s Beatles merch I saved years ago. And it delivers a great answer. I built a similar system to this a year ago with RAG but it was ass. A knowledge base that lets an agent find what it needs via a file system it actually understands just works better. The most magical thing now is as I add new things to my wiki (articles, images of inspo, meeting notes) the system will likely update 2-3 different articles where it feels that context belongs, or, just creates a new article. It's like this super genius librarian for your brain that's always filing stuff for your perfectly and also let's you easily query the knowledge for tasks useful to you (ex. design, product, writing, etc) and it never gets tired. I might spend next week productizing this, if that's of interest to you DM me + tell me your usecase!
Andrej Karpathy@karpathy

Wow, this tweet went very viral! I wanted share a possibly slightly improved version of the tweet in an "idea file". The idea of the idea file is that in this era of LLM agents, there is less of a point/need of sharing the specific code/app, you just share the idea, then the other person's agent customizes & builds it for your specific needs. So here's the idea in a gist format: gist.github.com/karpathy/442a6… You can give this to your agent and it can build you your own LLM wiki and guide you on how to use it etc. It's intentionally kept a little bit abstract/vague because there are so many directions to take this in. And ofc, people can adjust the idea or contribute their own in the Discussion which is cool.

English
212
338
4.3K
1.5M
Prof Adebayo
Prof Adebayo@ProfAdebay·
@FarzaTV honest question: why didn't you used RAG for it? Though this method of fetching data seems faster but i think it might not be effective for large files
English
0
0
1
41
Qwen
Qwen@Alibaba_Qwen·
Qwen3.6-Plus ranks # 1 on @OpenRouter , and the first model on OpenRouter to break 1 Trillion tokens processed in a single day！!🥇🔥 We are thrilled to see Qwen3.6-Plus topping the charts so quickly. This milestone wouldn't be possible without our amazing developers. ❤️Thank you！！
OpenRouter@OpenRouter

Qwen 3.6 Plus from @Alibaba_Qwen is officially the first model on OpenRouter to break 1 Trillion tokens processed in a single day! At ~1,400,000,000,000 tokens, it’s the strongest full day performance of any new model dropped this year. Congrats to the Qwen team!

English
95
119
1.5K
134.4K
Prof Adebayo
Prof Adebayo@ProfAdebay·
That’s the way forward, hopefully it gives their users more compute headroom to breathe. Anthropic is essentially prioritizing how its resources are allocated, even if that means turning down certain user demands and directing them to other models. It’s a calculated move, they understand their positioning in the market and are managing access in a way that maintains performance and exclusivity. One thing is certain: a large portion of the users they’re redirecting will likely keep their subscriptions, even if they can’t fully useit with Openclaw.
English
0
0
0
208
Yuchen Jin
Yuchen Jin@Yuchenj_UW·
Claude is cutting off apps like OpenClaw from using Claude subscriptions. I read somewhere that a $200/month Claude plan can use up to $5,000 in compute, so it’s heavily subsidized. Given Claude’s uptime issues, this might be the right move under current Anthropic GPU constraints. Codex is the more generous one for 3rd-party apps (OpenAI has more GPUs). It’ll be interesting to see how this strategy difference plays out.
Yuchen Jin tweet media
English
66
21
534
52K

Discover

@jenzhuscott @karpathy @iamZenderock @heynavtoor @theo @chrisparkX @Bent302 @0xSero