Video wird geladen...

Video konnte nicht geladen werden

Beim Laden dieses Videos ist ein Problem aufgetreten. Dies könnte an einem vorübergehenden Netzwerkproblem liegen oder das Video ist möglicherweise nicht verfügbar.

🚨 introducing hobbylm-1b computer use -- fully offline, open weights & we made it 13x faster: 3.4 s → 0.26 s per decision. my 1b model booked a call for sunday 7pm while wi‑fi was off. it found the name in contacts, worked out which sunday, set the time... in my time zone, asked y/n, then added the event to calendar with no internet and no cloud, on a macbook air. what else it does live offline as per tested: "split 600 among 4 people" → 150 in calculator "make my screen bright" → brightness up "sound is harsh, turn it down a notch" → volume down "mute everything" → muted "turn on dark mode" → dark mode on "bring back the light theme" → light mode on "i need to write something down" → notes opens "need papers from documents" → documents folder opens "seven minute timer for eggs" → 7-minute timer set "grab a screenshot" → saved to the desktop "what's the weather tomorrow" → "i can't do that one offline." yeah, how a 1b model covers this much in offline: 14 mb adapter on our own pretrained base, with a bit of fine-tuning, run fully on-device. thanks HARIHARAN M S for pulling this off weights, code and every test result:show more

Prajhan

1,539 subscribers

24,455 Aufrufe • vor 1 Tag •via X (Twitter)

Wissenschaft & Technologie Bildung

Anya Rossi• Live Now

Private livecam show

14 Kommentare

Profilbild von Rahul Kashyap

Rahul Kashyapvor 1 Tag

@hariharanms95 is cooking

Profilbild von Kevin Rajan

Kevin Rajanvor 1 Tag

@trycua @francedot 👀 this looks promising, gonna run some evals downstream for this haha

Profilbild von Prajhan

Prajhanvor 1 Tag

@trycua @francedot pls do

Profilbild von The Singularity Project

The Singularity Projectvor 1 Tag

Booking a calendar event with the Wi-Fi off, on a MacBook Air, from a 14 MB adapter. That's the future I want: personal, private, on-device Human-Machine Intelligence that runs your own machine with no cloud in the loop. I also like that it says "I can't do that one offline" instead of faking it. Great work, you two.

Profilbild von Eli Serrano

Eli Serranovor 1 Tag

Love the sealed-test discipline. Each round patched failures from the last one. Have you tried LLM-paraphrased requests or holding out whole phrasing families, to see if it generalizes past the templates rather than chasing them?

Profilbild von Thomas Linden

Thomas Lindenvor 1 Tag

Need to put this on a chip with internet and absolutely mogg the home assistant market

Profilbild von Sonam R

Sonam Rvor 1 Tag

Great job would definitely test this out. Probably have to make some changes for windows but a much needed tool

Profilbild von Mathias Heinke

Mathias Heinkevor 1 Tag

any chance to have this in German? :D

Profilbild von Vasant

Vasantvor 1 Tag

Available via Android app like run anywhere,atomic chat or else?

Profilbild von REAL ART IS MADE WITH AI

REAL ART IS MADE WITH AIvor 1 Tag

Can this run in Hermes? How good is it compared to Codex computer use?

Profilbild von niels .1

niels .1vor 1 Tag

two people, 60 h100 minutes, wifi off. the labs can keep their clusters

Profilbild von Daniel Baker

Daniel Bakervor 1 Tag

so damn cool! there's only one thing keeping me from trying to use it everywhere and that's the refusal rate. if it had 0% false positives you could reliably default to Hobby and then upgrade models whenever it fails

Profilbild von Solomon

Solomonvor 1 Tag

Sub-second decision loops on a 1B model running locally on an Air is wild. With OS action spaces, tiny models usually get tripped up by multi-step state verification when a dialog popup changes focus. Does the adapter rely on structured accessibility tree tokens or pure visual coordinate prediction?

Profilbild von bots stop following me

bots stop following mevor 1 Tag

Works on windows?

Ähnliche Videos

I run Claude Code with Qwen3 Coder Flash locally on my MacBook Air. It works offline, zero cloud, zero internet. No limit with all tokens on the house. Not great, not terrible- adequate performance for an on device AI agent chewing through code on a 1.24 kg laptop.

I run Claude Code with Qwen3 Coder Flash locally on my MacBook Air. It works offline, zero cloud, zero internet. No limit with all tokens on the house. Not great, not terrible- adequate performance for an on device AI agent chewing through code on a 1.24 kg laptop.

Mitko Vasilev

211,766 Aufrufe • vor 1 Jahr

Made a channel in my show to enhance the “ring” of the snare when I want to bring it back in for cool or heavy parts and put it on a mute so I can turn it on/off part to part. Pretty absurd EQ to get there but it’s sick in context. Keg mode dodgeball snare activated.

Made a channel in my show to enhance the “ring” of the snare when I want to bring it back in for cool or heavy parts and put it on a mute so I can turn it on/off part to part. Pretty absurd EQ to get there but it’s sick in context. Keg mode dodgeball snare activated.

THOTXBLASTER

20,690 Aufrufe • vor 2 Jahren

Alright guys, wow. Qwen 3.8 27B is basically Qwen 4. What we have here is one of the best one-shot shark survival games I've made with any model. I ran this same prompt through fable, and the biggest thing all models struggle with in this test is the top-down view of the sharks. Fable even had all of the fins reversed and details jumbled on the first turn. Before now, it was really a 2 to 8 turn test to get something final regardless of model. Qwen 3.8 27B just nailed this in one output, no harness at all, just purely the model thoroughly thinking before outputting the final result. On every asset used, different shark variants, everything, it all looks incredibly tight. The boat movement, shark speed and engagement, everything is so much better than I expected, and I was expecting a lot. Imagine what 8 turns would do in a harness? This confirms to me that while Qwen 3.8 27B thinks a lot, for the first time in a local model on the base weights, IT MIGHT ACTUALLY BE WORTH IT This is a Frontier Lab model result, and it ran on my 5090 at 60tps in Q5_K_M. I am just in complete astoundment at what Qwen has accomplished here, and I am incredibly grateful that this model is open source. The speed will also increase drastically with optimisations. This is just an incredible day. What a blessing. More to come!

Alright guys, wow. Qwen 3.8 27B is basically Qwen 4. What we have here is one of the best one-shot shark survival games I've made with any model. I ran this same prompt through fable, and the biggest thing all models struggle with in this test is the top-down view of the sharks. Fable even had all of the fins reversed and details jumbled on the first turn. Before now, it was really a 2 to 8 turn test to get something final regardless of model. Qwen 3.8 27B just nailed this in one output, no harness at all, just purely the model thoroughly thinking before outputting the final result. On every asset used, different shark variants, everything, it all looks incredibly tight. The boat movement, shark speed and engagement, everything is so much better than I expected, and I was expecting a lot. Imagine what 8 turns would do in a harness? This confirms to me that while Qwen 3.8 27B thinks a lot, for the first time in a local model on the base weights, IT MIGHT ACTUALLY BE WORTH IT This is a Frontier Lab model result, and it ran on my 5090 at 60tps in Q5_K_M. I am just in complete astoundment at what Qwen has accomplished here, and I am incredibly grateful that this model is open source. The speed will also increase drastically with optimisations. This is just an incredible day. What a blessing. More to come!

Kyle Hessling

26,198 Aufrufe • vor 2 Monaten

oh my.. this shouldn't be possible a 1B model that runs inside your browser, beats every model its size, and comes with its own desktop pet. MiniCPM-5 1B just changed the game for on-device AI. here's everything you need to know 🧵

oh my.. this shouldn't be possible a 1B model that runs inside your browser, beats every model its size, and comes with its own desktop pet. MiniCPM-5 1B just changed the game for on-device AI. here's everything you need to know 🧵

Muhammad Ayan

62,050 Aufrufe • vor 4 Monaten

You don't need to know how to code to turn a messy spreadsheet into something you can actually use. I had an Excel file of 20 brands I want to pitch. I dropped it in a folder, opened Qoder, and typed one sentence: turn this into a dashboard with search. Then I watched. It read the file, made a plan, wrote the code, opened the page in the browser, and tested the search on its own. I added a /goal so it knew when it was done: every brand shows, search works. The model: Qwen3.8-Flash, free through September 30. Zero Credits for this whole run. Want to try it? Enter my code BPU7C6 at first, then download Qoder and sign in. You will get 600 Credits: 300 from the trial and 300 from my code.

Jay

143,443 Aufrufe • vor 1 Monat

Jordi says Paramount+'s first UFC stream was "materially worse" than the pay-per-view model: Not only were the fighter walkouts replaced by 200 seconds of straight ads, but I don't think Paramount was ready for the influx of new live viewers. We were trying to watch the fights - it would play for a minute, then throw up a "fatal error." You had to turn the app off and turn it back on again. We did this for a while, then realized we could watch the Portuguese stream, so we ended up watching most of it with low volume in Portuguese.

Jordi says Paramount+'s first UFC stream was "materially worse" than the pay-per-view model: Not only were the fighter walkouts replaced by 200 seconds of straight ads, but I don't think Paramount was ready for the influx of new live viewers. We were trying to watch the fights - it would play for a minute, then throw up a "fatal error." You had to turn the app off and turn it back on again. We did this for a while, then realized we could watch the Portuguese stream, so we ended up watching most of it with low volume in Portuguese.

TBPN

35,432 Aufrufe • vor 9 Monaten

Nothing beats open-source! We now have a fully open-source coding agent that can compete with any proprietary solution out there: • 500+ available models • It doesn't train on your code • It doesn't charge a penny for tokens • It's open. You can change if you want. Kilo Code is the #1 app on OpenRouter right now, and the community is releasing new features on GitHub at rocket speed (about 20 releases in the last 30 days). I installed it in Visual Studio Code, but you can install it in pretty much any IDE. I need to spend more time with it, but here is a very interesting design choice: Most coding agents support two modes: an "agentic" mode for writing code and an "ask" mode for talking to the model without making any changes. Kilo Code supports 5 different modes: 1. Architect mode - Research, design, and planning 2. Debug mode - Problem solving and troubleshooting 3. Code mode - To write code. 4. Ask mode - To talk to the model. 5. Orchestrator mode - Combine every agent to solve problems. On paper, I really like this. I'll need more time to try out the specialized agents and see whether niching becomes an advantage. Here is the link to the website: Thanks to the Kilo (acq. by Anaconda) for partnering with me on this post.

Nothing beats open-source! We now have a fully open-source coding agent that can compete with any proprietary solution out there: • 500+ available models • It doesn't train on your code • It doesn't charge a penny for tokens • It's open. You can change if you want. Kilo Code is the #1 app on OpenRouter right now, and the community is releasing new features on GitHub at rocket speed (about 20 releases in the last 30 days). I installed it in Visual Studio Code, but you can install it in pretty much any IDE. I need to spend more time with it, but here is a very interesting design choice: Most coding agents support two modes: an "agentic" mode for writing code and an "ask" mode for talking to the model without making any changes. Kilo Code supports 5 different modes: 1. Architect mode - Research, design, and planning 2. Debug mode - Problem solving and troubleshooting 3. Code mode - To write code. 4. Ask mode - To talk to the model. 5. Orchestrator mode - Combine every agent to solve problems. On paper, I really like this. I'll need more time to try out the specialized agents and see whether niching becomes an advantage. Here is the link to the website: Thanks to the Kilo (acq. by Anaconda) for partnering with me on this post.

Santiago

73,304 Aufrufe • vor 11 Monaten

1/ Offline computer-use models are here. Today, one running on our Mac made it all the way to hotel checkout. We ran Qwen3.8-27B from Qwen through Cua Driver. It also turned a flyer into a Calendar event and reconciled receipts. Every model token stayed on the Mac 🧵

1/ Offline computer-use models are here. Today, one running on our Mac made it all the way to hotel checkout. We ran Qwen3.8-27B from Qwen through Cua Driver. It also turned a flyer into a Calendar event and reconciled receipts. Every model token stayed on the Mac 🧵

Francesco

27,540 Aufrufe • vor 2 Monaten

$KNX : Offline Notes. A New Primitive for Privacy. Architecture Explainer. Let us explore how offline transfers and note-based value work in KnoxNet. Instead of relying on a live network, value is converted into offline notes that exist on your device and move directly between users. Each note carries its own lineage — a verifiable history of ownership — secured through advanced encryption, enabling trustless transfers in a fully offline environment. This system is designed to operate independently at the moment of execution. No servers. No connectivity. No global state required. Watch the video to see how the world’s first offline-first cryptocurrency works. 👇

$KNX : Offline Notes. A New Primitive for Privacy. Architecture Explainer. Let us explore how offline transfers and note-based value work in KnoxNet. Instead of relying on a live network, value is converted into offline notes that exist on your device and move directly between users. Each note carries its own lineage — a verifiable history of ownership — secured through advanced encryption, enabling trustless transfers in a fully offline environment. This system is designed to operate independently at the moment of execution. No servers. No connectivity. No global state required. Watch the video to see how the world’s first offline-first cryptocurrency works. 👇

KnoxNet

21,921 Aufrufe • vor 5 Monaten

Announcing Tana offline 🔌🎉 (just in time to go off-grid this weekend!) We're super excited to launch our most-requested feature, offline mode! Not “select which pages to download” offline. Full knowledge graph offline. Whether you are AWS's US-EAST-1 server or a knowledge worker reliant on Aussie broadband, sometimes you just need to disconnect 🔌 🧵 So here’s the quick what’s-what on working offline:

Announcing Tana offline 🔌🎉 (just in time to go off-grid this weekend!) We're super excited to launch our most-requested feature, offline mode! Not “select which pages to download” offline. Full knowledge graph offline. Whether you are AWS's US-EAST-1 server or a knowledge worker reliant on Aussie broadband, sometimes you just need to disconnect 🔌 🧵 So here’s the quick what’s-what on working offline:

Tana

27,525 Aufrufe • vor 11 Monaten

I spent today turning a blank Hermes install into a sovereign local agent on my RTX 5090. always-on, reachable over telegram, fully local. here's everything it can do. it runs Qwen3.6-27B on the 5090 and answers from my phone over telegram. ~~~ I first made it faster. switched to the MTP build for self-speculative decoding: 62 > 115 tokens/sec. ~~~ it now benchmarks models on command. I ask it to speed-test a gguf and it stops its own model server to free the gpu, runs llama-bench, renders a card, sends it to my phone, then restarts itself. it's never left offline. ~~~ it also knows me very well. I pointed it to my private HF repo where I store all my AI traces since I started using it. I created a local markdown memory vault, seeded from my own history, with semantic search over it. ~~~ the whole stack stays home: weights on disk, inference on the 5090, the agent in my pocket. here is the video it created to showcase everything it has now.

I spent today turning a blank Hermes install into a sovereign local agent on my RTX 5090. always-on, reachable over telegram, fully local. here's everything it can do. it runs Qwen3.6-27B on the 5090 and answers from my phone over telegram. ~ I first made it faster. switched to the MTP build for self-speculative decoding: 62 > 115 tokens/sec. ~ it now benchmarks models on command. I ask it to speed-test a gguf and it stops its own model server to free the gpu, runs llama-bench, renders a card, sends it to my phone, then restarts itself. it's never left offline. ~ it also knows me very well. I pointed it to my private HF repo where I store all my AI traces since I started using it. I created a local markdown memory vault, seeded from my own history, with semantic search over it. ~ the whole stack stays home: weights on disk, inference on the 5090, the agent in my pocket. here is the video it created to showcase everything it has now.

witcheer

15,041 Aufrufe • vor 4 Monaten

Hunter Schafer admits she got cheated on for the first time after a lifetime of polyamory and it fundamentally changed her "Everything else had been polyamorous. That’s what I came up in. That’s how I learned relationships." "I looked at people in monogamy and thought you’re cheating on yourself." "This was my first monogamous relationship with a cis man." "I got cheated on for the first time and it fundamentally changed me." "Cheating has nothing to do with you. It has everything to do with that person and whatever pain they’re in." "It opens up the door for that kind of betrayal which is a different world of pain."

Hunter Schafer admits she got cheated on for the first time after a lifetime of polyamory and it fundamentally changed her "Everything else had been polyamorous. That’s what I came up in. That’s how I learned relationships." "I looked at people in monogamy and thought you’re cheating on yourself." "This was my first monogamous relationship with a cis man." "I got cheated on for the first time and it fundamentally changed me." "Cheating has nothing to do with you. It has everything to do with that person and whatever pain they’re in." "It opens up the door for that kind of betrayal which is a different world of pain."

mercante

2,458,346 Aufrufe • vor 2 Monaten

Apple built a large foundation model and fine-tuned it on multiple tasks. But they are doing something very clever: They load a single model in memory and use different adapters to specialize the model on the fly. I recorded a video to show you how to write the code to do the same thing Apple is doing. I explain everything step by step. Here is what I'll show you in the video: 1. We'll load two datasets 2. Then load a large model 3. Then, we'll fine-tune the model on both datasets I'll use LoRA to fine-tune the model. This process creates two small adapters, each specializing in solving one of the datasets. The base model's original parameters will remain unchanged. From here: 4. We'll generate a list of tasks 5. We'll load the correct adapter to solve each task The large model I'm using needs 346 MB of memory, but I only need to load it once. Each adapter is only 2.7 MB. I only need to load the base model once and pair it with any of the fine-tuned adapters. Minimum memory footprint and I can solve multiple tasks. Hope this helps!

Apple built a large foundation model and fine-tuned it on multiple tasks. But they are doing something very clever: They load a single model in memory and use different adapters to specialize the model on the fly. I recorded a video to show you how to write the code to do the same thing Apple is doing. I explain everything step by step. Here is what I'll show you in the video: 1. We'll load two datasets 2. Then load a large model 3. Then, we'll fine-tune the model on both datasets I'll use LoRA to fine-tune the model. This process creates two small adapters, each specializing in solving one of the datasets. The base model's original parameters will remain unchanged. From here: 4. We'll generate a list of tasks 5. We'll load the correct adapter to solve each task The large model I'm using needs 346 MB of memory, but I only need to load it once. Each adapter is only 2.7 MB. I only need to load the base model once and pair it with any of the fine-tuned adapters. Minimum memory footprint and I can solve multiple tasks. Hope this helps!

Santiago

84,747 Aufrufe • vor 2 Jahren

Nobody seems to know how insane GPT-5.4 is with computer use. I asked GPT-5.4 to draw the OpenAI logo in Microsoft Paint. No computer use API. Just a screenshot and basic tool calls (click, drag, press_key) all coordinate-based. The first drawing was awful. And GPT knew it. It looked at its own result and essentially went "yeah, no." What happened next is what broke my brain: It opened a browser. Went to Bing Images. Searched for the OpenAI logo. Found one. Then (and I cannot stress this enough) it used the Windows area screenshot shortcut (Win+Shift+S) to snip just the logo off the screen. Went back to Paint. Imported it. Centered it. All on its own. No instructions to do any of that. It just improvised a better strategy when the first one failed. My prompt was "Draw the OpenAI logo" with Paint already opened on the computer. Sure, it's "cheating." But honestly? That's exactly what I'd do too. And the fact that it came up with this plan from nothing but a screenshot and a coordinate system is wild.

Clad3815

1,070,585 Aufrufe • vor 7 Monaten

Last night, I asked my Builder Grok Bot to send a job to Cursor. I asked it to have Cursor create an “explainer” video pipeline that follows a set of specific instructions, and is easy to run whenever I want to create an educational video vs starting from scratch every time. My Grok Bot started up a Cursor project in the cloud. One Cursor coordinator agent ran the project and handed work to two others. The first working agent built the pipeline in the cloud. After it finished, it sent the work from the cloud down to a Cursor worker living in my Mac Mini. The Mac Mini worker installed all of the programs necessary for building these videos, and will follow these instructions for all future builds when I run this pipeline: 1. Claude Code on the Mini, signed in to my Claude Pro subscription, writes the research, the script and the animation plan. 2. Python on the Mini draws the animations. 3. Kokoro on the Mini records the narration. 4. ffmpeg on the Mini adds the music and captions and exports the MP4. 5. The finished video saves to my iCloud folder. Now every time I ask my Grok Bot to make an educational video, it passes it through this pipeline on my Mac mini. All of the tools are already installed, and the steps are set in stone. All I need to do is give it a topic and it knows to run this. This all happens on my Mac Mini now, and doesn’t involve cloud agents unless I want to bring one on to edit the pipeline. No need to start from scratch every time. Here’s the first video explaining how the power grid works:

Last night, I asked my Builder Grok Bot to send a job to Cursor. I asked it to have Cursor create an “explainer” video pipeline that follows a set of specific instructions, and is easy to run whenever I want to create an educational video vs starting from scratch every time. My Grok Bot started up a Cursor project in the cloud. One Cursor coordinator agent ran the project and handed work to two others. The first working agent built the pipeline in the cloud. After it finished, it sent the work from the cloud down to a Cursor worker living in my Mac Mini. The Mac Mini worker installed all of the programs necessary for building these videos, and will follow these instructions for all future builds when I run this pipeline: 1. Claude Code on the Mini, signed in to my Claude Pro subscription, writes the research, the script and the animation plan. 2. Python on the Mini draws the animations. 3. Kokoro on the Mini records the narration. 4. ffmpeg on the Mini adds the music and captions and exports the MP4. 5. The finished video saves to my iCloud folder. Now every time I ask my Grok Bot to make an educational video, it passes it through this pipeline on my Mac mini. All of the tools are already installed, and the steps are set in stone. All I need to do is give it a topic and it knows to run this. This all happens on my Mac Mini now, and doesn’t involve cloud agents unless I want to bring one on to edit the pipeline. No need to start from scratch every time. Here’s the first video explaining how the power grid works:

Mike P

11,092 Aufrufe • vor 23 Tagen

Venue internet got you down? Try Offline Mode for Startgg by Nicolet! This offline-first client lets you see and report matches offline, then syncs online once internet is back. You can even grant per-pool access to report matches from other browsers on your local network!

Venue internet got you down? Try Offline Mode for Startgg by Nicolet! This offline-first client lets you see and report matches offline, then syncs online once internet is back. You can even grant per-pool access to report matches from other browsers on your local network!

start.gg

177,418 Aufrufe • vor 5 Monaten

Claude Code went from research preview to a $1B product in 6 months. Cowork is the same thing, built for people who don’t code. This free 18-minute video teaches you how to set it up from scratch. No code. No fluff. No $500 upsell. Watch it. Bookmark it. The few who actually do will run their entire week on autopilot:

Claude Code went from research preview to a $1B product in 6 months. Cowork is the same thing, built for people who don’t code. This free 18-minute video teaches you how to set it up from scratch. No code. No fluff. No $500 upsell. Watch it. Bookmark it. The few who actually do will run their entire week on autopilot:

Evan Luthra

31,725 Aufrufe • vor 5 Monaten

🚨 VWP SANDBOX now LIVE on View live VWP data from all NOAA radar sites + terminal sites. Explore them in 3D or 2D to understand more about how hodographs and dynamics work! Toggle on dark mode or light mode for your viewing pleasure, as well as download the geojson file of the VWP to save and come back to it later by loading it back in to the system! This works a lot better on computer but I tried my best at getting mobile at least somewhat usable. For more, go to the "How To Use" section.

🚨 VWP SANDBOX now LIVE on View live VWP data from all NOAA radar sites + terminal sites. Explore them in 3D or 2D to understand more about how hodographs and dynamics work! Toggle on dark mode or light mode for your viewing pleasure, as well as download the geojson file of the VWP to save and come back to it later by loading it back in to the system! This works a lot better on computer but I tried my best at getting mobile at least somewhat usable. For more, go to the "How To Use" section.

James Pettus

30,106 Aufrufe • vor 1 Jahr

WHY IS NO ONE TALKING ABOUT THIS?? Gemma 3n model was one of the best surprises for me. The fact that you can run it on edge devices even with just 2GB of RAM is impressive. A few weeks back, I was on holiday and used the Gemini Live feature a lot. But I kept running into issues whenever I was in a place where the network wasn’t reliable. Gemma 3n : > Multimodal: Supports text and image inputs; video/audio in development. > Context Window: Up to 128K tokens (32K for 1B model). > Multilingual: Trained on 140+ languages. > Privacy: Offline, on-device processing for data security. > Model Sizes: E2B (5B parameters, 2GB RAM) and E4B (10–12B parameters, 3GB RAM). Video credit: Google YT

WHY IS NO ONE TALKING ABOUT THIS?? Gemma 3n model was one of the best surprises for me. The fact that you can run it on edge devices even with just 2GB of RAM is impressive. A few weeks back, I was on holiday and used the Gemini Live feature a lot. But I kept running into issues whenever I was in a place where the network wasn’t reliable. Gemma 3n : > Multimodal: Supports text and image inputs; video/audio in development. > Context Window: Up to 128K tokens (32K for 1B model). > Multilingual: Trained on 140+ languages. > Privacy: Offline, on-device processing for data security. > Model Sizes: E2B (5B parameters, 2GB RAM) and E4B (10–12B parameters, 3GB RAM). Video credit: Google YT

AshutoshShrivastava

210,641 Aufrufe • vor 1 Jahr