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.

NVIDIA just made AI detect objects 10x faster by deleting one step. It's called LocateAnything, and it removes the biggest bottleneck no one else was fixing in vision-language models. Normally a model builds each bounding box one coordinate token at a time. 100 objects means thousands of tokens before... an answer. NVIDIA scrapped that: their Parallel Box Decoding predicts the whole box in a single forward pass, as one atomic unit. → 12.7 boxes/sec on one H100 → 10x faster than Qwen3-VL → +3.8% F1 on LVIS, accuracy up, not down → 3B params, runs on one consumer GPU Treating the box as one unit keeps its coordinates tied together, which is why accuracy climbed instead of falling. One model handles detection, GUI grounding, OCR, and document understanding, ready for computer-use agents, robotics, and document pipelines. 100% open source, weights, code, demo, and paper all live.show more

Alvaro Cintas

130,505 subscribers

162,747 Aufrufe • vor 1 Tag •via X (Twitter)

Bildung Nachrichten & Politik Wissenschaft & Technologie

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

💡 FM26 Quick Tip: Tick Multiple Boxes Faster Instead of ticking each box one by one, click on the first item you want, then hold Shift and click on the last one: everything in between will be selected automatically. ☑️

💡 FM26 Quick Tip: Tick Multiple Boxes Faster Instead of ticking each box one by one, click on the first item you want, then hold Shift and click on the last one: everything in between will be selected automatically. ☑️

Sort It Out SI.net

54,376 Aufrufe • vor 8 Monaten

THIS GUY BOUGHT A $2,400 NVIDIA BOX AND SAVED $18,700/YEAR ON CLOUD GPUS WITHOUT RENTING SERVERS AGAIN the entire setup runs on one rule - stop paying every time you want to test something most people run 20 small AI experiments in the cloud and think it’s cheap because each one looks harmless - then the invoice comes in and suddenly their “side project” has the same monthly cost as a car payment he made the same mistake for months and it slowly killed the way he worked one box, one desk, local models - and now he can run tests overnight without thinking about hourly GPU prices $18,700/year saved by a little NVIDIA box he can literally hold in his hands

THIS GUY BOUGHT A $2,400 NVIDIA BOX AND SAVED $18,700/YEAR ON CLOUD GPUS WITHOUT RENTING SERVERS AGAIN the entire setup runs on one rule - stop paying every time you want to test something most people run 20 small AI experiments in the cloud and think it’s cheap because each one looks harmless - then the invoice comes in and suddenly their “side project” has the same monthly cost as a car payment he made the same mistake for months and it slowly killed the way he worked one box, one desk, local models - and now he can run tests overnight without thinking about hourly GPU prices $18,700/year saved by a little NVIDIA box he can literally hold in his hands

Gipp 🦅

12,484 Aufrufe • vor 26 Tagen

A peanut-sized Chinese model just dethroned Gemini at reading documents. GLM-OCR is a 0.9B parameter vision-language model. It scores 94.62 on OmniDocBench V1.5, ranking #1 overall. For context, it outperforms models 100x its size. 100% open-source. It works in two stages. 1. A layout engine detects every region in a document. 2. Each region gets read in parallel. The model predicts multiple tokens per step instead of one. That's what makes it so fast at small size. It handles things most OCR tools struggle with: > Complex tables and nested layouts > Handwritten text and stamps > Math formulas and code blocks > Mixed image-and-text documents You can run it locally through Ollama. It fits on edge devices with limited compute. Every expensive OCR API just got a free competitor.

A peanut-sized Chinese model just dethroned Gemini at reading documents. GLM-OCR is a 0.9B parameter vision-language model. It scores 94.62 on OmniDocBench V1.5, ranking #1 overall. For context, it outperforms models 100x its size. 100% open-source. It works in two stages. 1. A layout engine detects every region in a document. 2. Each region gets read in parallel. The model predicts multiple tokens per step instead of one. That's what makes it so fast at small size. It handles things most OCR tools struggle with: > Complex tables and nested layouts > Handwritten text and stamps > Math formulas and code blocks > Mixed image-and-text documents You can run it locally through Ollama. It fits on edge devices with limited compute. Every expensive OCR API just got a free competitor.

AlphaSignal AI

91,821 Aufrufe • vor 2 Monaten

A peanut-sized Chinese model just dethroned Gemini at reading documents. GLM-OCR is a 0.9B parameter vision-language model. It scores 94.62 on OmniDocBench V1.5, ranking #1 overall. For context, it outperforms models 100x its size. 100% open-source. It works in two stages. 1. A layout engine detects every region in a document. 2. Each region gets read in parallel. The model predicts multiple tokens per step instead of one. That's what makes it so fast at small size. It handles things most OCR tools struggle with: > Complex tables and nested layouts > Handwritten text and stamps > Math formulas and code blocks > Mixed image-and-text documents You can run it locally through Ollama. It fits on edge devices with limited compute. Every expensive OCR API just got a free competitor.

A peanut-sized Chinese model just dethroned Gemini at reading documents. GLM-OCR is a 0.9B parameter vision-language model. It scores 94.62 on OmniDocBench V1.5, ranking #1 overall. For context, it outperforms models 100x its size. 100% open-source. It works in two stages. 1. A layout engine detects every region in a document. 2. Each region gets read in parallel. The model predicts multiple tokens per step instead of one. That's what makes it so fast at small size. It handles things most OCR tools struggle with: > Complex tables and nested layouts > Handwritten text and stamps > Math formulas and code blocks > Mixed image-and-text documents You can run it locally through Ollama. It fits on edge devices with limited compute. Every expensive OCR API just got a free competitor.

Jafar Najafov

13,630 Aufrufe • vor 2 Monaten

it's open source time, with a real leap for world models 🎉 NVIDIA's SANA-WM: a camera-conditioned world model that fits on one GPU. 60s of 720p in 34s on a single 5090 - 2.6B params and Apache 2.0!

it's open source time, with a real leap for world models 🎉 NVIDIA's SANA-WM: a camera-conditioned world model that fits on one GPU. 60s of 720p in 34s on a single 5090 - 2.6B params and Apache 2.0!

Victor M

34,386 Aufrufe • vor 1 Monat

Fine-tune DeepSeek-OCR on your own language! (100% local) DeepSeek-OCR is a 3B-parameter vision model that achieves 97% precision while using 10× fewer vision tokens than text-based LLMs. It handles tables, papers, and handwriting without killing your GPU or budget. Why it matters: Most vision models treat documents as massive sequences of tokens, making long-context processing expensive and slow. DeepSeek-OCR uses context optical compression to convert 2D layouts into vision tokens, enabling efficient processing of complex documents. The best part? You can easily fine-tune it for your specific use case on a single GPU. I used Unsloth to run this experiment on Persian text and saw an 88.26% improvement in character error rate. ↳ Base model: 149% character error rate (CER) ↳ Fine-tuned model: 60% CER (57% more accurate) ↳ Training time: 60 steps on a single GPU Persian was just the test case. You can swap in your own dataset for any language, document type, or specific domain you're working with. I've shared the complete guide in the next tweet - all the code, notebooks, and environment setup ready to run with a single click. Everything is 100% open-source!

Fine-tune DeepSeek-OCR on your own language! (100% local) DeepSeek-OCR is a 3B-parameter vision model that achieves 97% precision while using 10× fewer vision tokens than text-based LLMs. It handles tables, papers, and handwriting without killing your GPU or budget. Why it matters: Most vision models treat documents as massive sequences of tokens, making long-context processing expensive and slow. DeepSeek-OCR uses context optical compression to convert 2D layouts into vision tokens, enabling efficient processing of complex documents. The best part? You can easily fine-tune it for your specific use case on a single GPU. I used Unsloth to run this experiment on Persian text and saw an 88.26% improvement in character error rate. ↳ Base model: 149% character error rate (CER) ↳ Fine-tuned model: 60% CER (57% more accurate) ↳ Training time: 60 steps on a single GPU Persian was just the test case. You can swap in your own dataset for any language, document type, or specific domain you're working with. I've shared the complete guide in the next tweet - all the code, notebooks, and environment setup ready to run with a single click. Everything is 100% open-source!

Akshay 🚀

126,077 Aufrufe • vor 7 Monaten

One Aprilia is on pole, the other one has just broken down and Jorge Martín Almoguera's back in the box 👀 #SanMarinoGP 🇸🇲

One Aprilia is on pole, the other one has just broken down and Jorge Martín Almoguera's back in the box 👀 #SanMarinoGP 🇸🇲

MotoGP™🏁

25,748 Aufrufe • vor 9 Monaten

Qwen3.6 35B A3B can't fill out a paper form on its own. But give it NVIDIA's LocateAnything-3B — the #1 trending model on HuggingFace — as its eyes, and the two small models get it done together. (The test: place each element at the right pixel position on a blank form image, not type into a field.) Setup: > Qwen is the brain (main model), LocateAnything is the eyes (helper model acting as a tool). > I gave Qwen a new tool: ask "where's the email field?" and LocateAnything returns the exact x, y, width, height. > The blue boxes on the screen are its detections. Look how tight they are — it nails every field. Result: > Qwen3.6 35B A3B + LocateAnything-3B: form completed, all info correct. > Name, DOB, ID, gender, marital status, nationality, email, phone, address, postal code: all landed in the right field areas. > Character-box alignment still a touch loose, but every value is where it belongs. > 9m10s, 224.5k input, 24.3k output, 21 turns. Why it matters: > Qwen alone can't finish this test. Bolt on a 3B model that does exactly one thing > locate > and suddenly it can. > A combination of small models can do the work of a single large one.

Qwen3.6 35B A3B can't fill out a paper form on its own. But give it NVIDIA's LocateAnything-3B — the #1 trending model on HuggingFace — as its eyes, and the two small models get it done together. (The test: place each element at the right pixel position on a blank form image, not type into a field.) Setup: > Qwen is the brain (main model), LocateAnything is the eyes (helper model acting as a tool). > I gave Qwen a new tool: ask "where's the email field?" and LocateAnything returns the exact x, y, width, height. > The blue boxes on the screen are its detections. Look how tight they are — it nails every field. Result: > Qwen3.6 35B A3B + LocateAnything-3B: form completed, all info correct. > Name, DOB, ID, gender, marital status, nationality, email, phone, address, postal code: all landed in the right field areas. > Character-box alignment still a touch loose, but every value is where it belongs. > 9m10s, 224.5k input, 24.3k output, 21 turns. Why it matters: > Qwen alone can't finish this test. Bolt on a 3B model that does exactly one thing > locate > and suddenly it can. > A combination of small models can do the work of a single large one.

stevibe

148,262 Aufrufe • vor 26 Tagen

This uniquely created deck box, our Soft Crate 80+, is the perfect all-in-one solution to transport your deck and tokens! This sturdy and compact box has space for 80 double-sleeved cards or 100 single-sleeved cards & comes with an innovative token box and divider included! (1/2)

This uniquely created deck box, our Soft Crate 80+, is the perfect all-in-one solution to transport your deck and tokens! This sturdy and compact box has space for 80 double-sleeved cards or 100 single-sleeved cards & comes with an innovative token box and divider included! (1/2)

Gamegenic

14,170 Aufrufe • vor 9 Monaten

Tonight! and now on BBC iPlayer - if you're not one of the millions who've already watched it, you can now do all 3 in a box set binge - State of Chaos box set

Tonight! and now on BBC iPlayer - if you're not one of the millions who've already watched it, you can now do all 3 in a box set binge - State of Chaos box set

Laura Kuenssberg

579,967 Aufrufe • vor 2 Jahren

1/ NVIDIA just open-sourced Cosmos 3 at GTC Taipei! It's the first fully open "omnimodel" for physical AI - one model that understands the real world, predicts what happens next, and generates the actions a robot should take. Weights, code, datasets. All open. And this is really big. Lets dig into everything: 🧵

1/ NVIDIA just open-sourced Cosmos 3 at GTC Taipei! It's the first fully open "omnimodel" for physical AI - one model that understands the real world, predicts what happens next, and generates the actions a robot should take. Weights, code, datasets. All open. And this is really big. Lets dig into everything: 🧵

Chubby♨️

18,067 Aufrufe • vor 27 Tagen

JENSEN HUANG UNVEILED A BOARD THAT RUNS 1 TRILLION PARAMETER AI MODELS. THE $249 NVIDIA BOX UNDER YOUR DESK KILLS A $200/MONTH AI BILL FOR $5 IN ELECTRICITY jensen held it up on stage with one hand and called it the architecture that runs the future of ai. that same technology now ships in a $249 box smaller than your wallet the jetson orin nano super pulls 7-25 watts and does 67 trillion ai operations per second. llama 3, mistral and deepseek run locally with no api fees and no data leaving your machine most developers pay $2,400 a year across chatgpt, openai api, claude pro and cursor. the jetson costs $314 in year one and $60 a year after. 2 year savings hit $4,431 install ollama with one command, change one line of code to point at localhost, and every tool built for openai works identically. zero rewrites, zero rate limits cloud subscriptions keep getting more expensive and rate limits keep getting tighter. the people who own the box in 2026 are going to look very far ahead in 2028 bookmark this and read the article below

JENSEN HUANG UNVEILED A BOARD THAT RUNS 1 TRILLION PARAMETER AI MODELS. THE $249 NVIDIA BOX UNDER YOUR DESK KILLS A $200/MONTH AI BILL FOR $5 IN ELECTRICITY jensen held it up on stage with one hand and called it the architecture that runs the future of ai. that same technology now ships in a $249 box smaller than your wallet the jetson orin nano super pulls 7-25 watts and does 67 trillion ai operations per second. llama 3, mistral and deepseek run locally with no api fees and no data leaving your machine most developers pay $2,400 a year across chatgpt, openai api, claude pro and cursor. the jetson costs $314 in year one and $60 a year after. 2 year savings hit $4,431 install ollama with one command, change one line of code to point at localhost, and every tool built for openai works identically. zero rewrites, zero rate limits cloud subscriptions keep getting more expensive and rate limits keep getting tighter. the people who own the box in 2026 are going to look very far ahead in 2028 bookmark this and read the article below

starmex

54,309 Aufrufe • vor 27 Tagen

🇮🇹 WORLD'S THINNEST CAR IS SKINNIER THAN A MOTORBIKE An Italian creator just built a real-life cartoon car that’s only 19 inches wide, has one headlight and paper-thin tires. It’s called the Panda for One, rolls on four wheels, has just one seat (maybe two if you’re tiny), and looks flatter than a pizza box. Even its mirrors stick out wider than the car itself. Source: The Sun, Luxurious

🇮🇹 WORLD'S THINNEST CAR IS SKINNIER THAN A MOTORBIKE An Italian creator just built a real-life cartoon car that’s only 19 inches wide, has one headlight and paper-thin tires. It’s called the Panda for One, rolls on four wheels, has just one seat (maybe two if you’re tiny), and looks flatter than a pizza box. Even its mirrors stick out wider than the car itself. Source: The Sun, Luxurious

Mario Nawfal

299,111 Aufrufe • vor 11 Monaten

Remaking all the SM64 objects : Goombas! This new model has 37 fewer tris, and only uses 2 materials, one opaque and one transparent, which allows all Goombas to be rendered in a single draw call. But the biggest optimisation comes from fixing Nintendo's mistakes... 🧵

Remaking all the SM64 objects : Goombas! This new model has 37 fewer tris, and only uses 2 materials, one opaque and one transparent, which allows all Goombas to be rendered in a single draw call. But the biggest optimisation comes from fixing Nintendo's mistakes... 🧵

Kaze Emanuar

216,463 Aufrufe • vor 1 Jahr

Trained a simple world model for my robot arm. It predicts the future over 20000 times faster than real time on a single NVIDIA RTX 3090 GPU (128 batch -> 160x faster each).

Trained a simple world model for my robot arm. It predicts the future over 20000 times faster than real time on a single NVIDIA RTX 3090 GPU (128 batch -> 160x faster each).

Alexander Koch

39,700 Aufrufe • vor 2 Jahren

Diffusion for virtual worlds: we've trained a new model to create 3D objects from text. And it's 50x faster than any alternative. We made each of these cute characters in just one minute with Genmo's new text-to-3D model. Build your own world at

Diffusion for virtual worlds: we've trained a new model to create 3D objects from text. And it's 50x faster than any alternative. We made each of these cute characters in just one minute with Genmo's new text-to-3D model. Build your own world at

Genmo

18,528 Aufrufe • vor 3 Jahren

While shopping at Aldi, a woman picked up a box of lettuce—only to discover a live frog inside. Maybe the store was running a “Buy one box of lettuce, get one frog free” special! 🤣🤣🤣

While shopping at Aldi, a woman picked up a box of lettuce—only to discover a live frog inside. Maybe the store was running a “Buy one box of lettuce, get one frog free” special! 🤣🤣🤣

I Meme Therefore I Am 🇺🇸

113,143 Aufrufe • vor 11 Monaten

Introducing: Nvidia Pro Mode! Basically GPT-5 Pro, but for the new Nvidia open source model :) Pro Mode chains together up to 50 instances of the new Nemotron 9B model, enabling it to produce a better answer than one instance alone could create! Open source, link here:

Introducing: Nvidia Pro Mode! Basically GPT-5 Pro, but for the new Nvidia open source model :) Pro Mode chains together up to 50 instances of the new Nemotron 9B model, enabling it to produce a better answer than one instance alone could create! Open source, link here:

Matt Shumer

292,390 Aufrufe • vor 9 Monaten

OF model Bonnie Blue brags about her plans to sleep with 2,000 men in one day and be “tied up in a glass box and open to the public.” Blue has told men to “get involved or just walk around and watch” as she plans the shocking 2,000 man stunt.

Sensitive content

OF model Bonnie Blue brags about her plans to sleep with 2,000 men in one day and be “tied up in a glass box and open to the public.” Blue has told men to “get involved or just walk around and watch” as she plans the shocking 2,000 man stunt.

Oli London

13,602,349 Aufrufe • vor 1 Jahr

Getting a new TV for Christmas? Put your old broken one in the box of the new one and watch a porch pirate take it for free. Follow me for more life hacks.

Getting a new TV for Christmas? Put your old broken one in the box of the new one and watch a porch pirate take it for free. Follow me for more life hacks.

Planet Of Memes

638,292 Aufrufe • vor 6 Monaten