正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

We have started taking DYNA-1, our dexterous robust VLA model, to conferences and showcasing it for hours on end! The model run for 3 days, 8 hours each day at #HITEC2025 3 weeks ago with 99.9% overall success rate (dropped 1 towel in day 2). No intervention, it just... show more

Dyna Robotics

6,803 subscribers

55,390 次观看 • 1 年前 •via X (Twitter)

科学技术 #HITEC2025

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

This is what Dubai Airport Terminal 3 has looked like for the last 4 days. Day and night. - Our flight kept getting delayed and cancelled for THREE days. - After each cancellation, you have to rebook in queues like this. - I stood for two nights straight. EIGHT HOURS each time.

This is what Dubai Airport Terminal 3 has looked like for the last 4 days. Day and night. - Our flight kept getting delayed and cancelled for THREE days. - After each cancellation, you have to rebook in queues like this. - I stood for two nights straight. EIGHT HOURS each time.

Solomon King #FreeBesigye

53,242 次观看 • 2 年前

GPT-5.5 is MUCH more reliable on longer running tasks - for the first time with any model. As we speak I have a migration running for over 7+ hours - this literally never happened before, the models would maybe run for 30 mins or of you really shout at them for 2-3 hours. Last night I went to sleep, set a long running task, then queued up 10 prompts to 'keep it going'. It did not stop after the first prompt and kept going for 8+ hours and I woke up to all the same prompts still queued up. The ability to run for a long time, in combination with ability to validate with computer use & other tools, makes it much more useful for building real applications.

GPT-5.5 is MUCH more reliable on longer running tasks - for the first time with any model. As we speak I have a migration running for over 7+ hours - this literally never happened before, the models would maybe run for 30 mins or of you really shout at them for 2-3 hours. Last night I went to sleep, set a long running task, then queued up 10 prompts to 'keep it going'. It did not stop after the first prompt and kept going for 8+ hours and I woke up to all the same prompts still queued up. The ability to run for a long time, in combination with ability to validate with computer use & other tools, makes it much more useful for building real applications.

Peter Gostev

105,642 次观看 • 3 个月前

We finally pulled the plug on per day pricing. It had to be done. The problem with per day pricing is it always creates winners and losers. - If we’re on site for 8 hours and don’t quite finish, we feel obligated to work extra to get it done. We lose. - If we finish a project in 5 hours and the customer paid for 8, they feel like they lost. It’s a simple system to start with. Customers understand it, it’s easy to quote, but it’s not scalable. It relies too much on experience and too much back and forth. So we switched to project based pricing. Now we’ve got 3 tiers of brush density and 3 tiers of terrain. Tier 1 is easy, Tier 3 is tough. Pricing scales per quarter acre, with volume discounts on larger jobs. It seems more complex on paper, but in reality it’s repeatable, reliable, and something we can train any new hire to do. We calculate the project, send the estimate, and that’s it. No more “estimated days” language. No more winners and losers. We’re much happier with this system.

We finally pulled the plug on per day pricing. It had to be done. The problem with per day pricing is it always creates winners and losers. - If we’re on site for 8 hours and don’t quite finish, we feel obligated to work extra to get it done. We lose. - If we finish a project in 5 hours and the customer paid for 8, they feel like they lost. It’s a simple system to start with. Customers understand it, it’s easy to quote, but it’s not scalable. It relies too much on experience and too much back and forth. So we switched to project based pricing. Now we’ve got 3 tiers of brush density and 3 tiers of terrain. Tier 1 is easy, Tier 3 is tough. Pricing scales per quarter acre, with volume discounts on larger jobs. It seems more complex on paper, but in reality it’s repeatable, reliable, and something we can train any new hire to do. We calculate the project, send the estimate, and that’s it. No more “estimated days” language. No more winners and losers. We’re much happier with this system.

Alex B

25,944 次观看 • 11 个月前

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

92,002 次观看 • 4 个月前

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 次观看 • 3 个月前

What have they done to Battlefield ?? What have they laced it with? >Play 4 hours >Get bored of the map rotation and no big maps >That's probably enough Battlefield for the day >30 mins go by >Start missing it >Load it up >Play another 4 hours >Repeat 3 times per day

What have they done to Battlefield ?? What have they laced it with? >Play 4 hours >Get bored of the map rotation and no big maps >That's probably enough Battlefield for the day >30 mins go by >Start missing it >Load it up >Play another 4 hours >Repeat 3 times per day

butch.

1,183,739 次观看 • 11 个月前

Day 8. START! As promised, we have an insane day. Bringing all the years together into a 3D model. ⚠️ TODAY AT 1 PM CT ⚠️ Credits to Yuniiho and Nanoless_ for the design and making this possible. Special thanks to Mari_Art and Baffler for this. 🔥🐉

Day 8. START! As promised, we have an insane day. Bringing all the years together into a 3D model. ⚠️ TODAY AT 1 PM CT ⚠️ Credits to Yuniiho and Nanoless_ for the design and making this possible. Special thanks to Mari_Art and Baffler for this. 🔥🐉

Zentreya⚡🐲⚔️ SVA

76,055 次观看 • 1 年前

And here we go — welcome to Day 1 of Based 7 Days of Xmas. We’re starting simple… but essential. 🎁 Day 1 Challenge 1) Quote tweet this post 2) Tag a fellow Pal & wish them “Merry Xmas from all of us at Based” in your native language 3) Use #BASEDXMAS 🎲 3 winners chosen at random ❤️ But if an entry truly stands out… it might just win our hearts and who knows.. it is the season of giving after all ;) 🎁 Presents to be won 🥇 1st: 6 Blind Boxes 🥈 2nd: 4 Blind Boxes 🥉 3rd: 2 Blind Boxes Merry Day 1 Xmas Pals 🎄✨ Let the cheers begin! ⏳24 hours

And here we go — welcome to Day 1 of Based 7 Days of Xmas. We’re starting simple… but essential. 🎁 Day 1 Challenge 1) Quote tweet this post 2) Tag a fellow Pal & wish them “Merry Xmas from all of us at Based” in your native language 3) Use #BASEDXMAS 🎲 3 winners chosen at random ❤️ But if an entry truly stands out… it might just win our hearts and who knows.. it is the season of giving after all ;) 🎁 Presents to be won 🥇 1st: 6 Blind Boxes 🥈 2nd: 4 Blind Boxes 🥉 3rd: 2 Blind Boxes Merry Day 1 Xmas Pals 🎄✨ Let the cheers begin! ⏳24 hours

BasedPals

24,022 次观看 • 7 个月前

Our development team is breaking new ground in market prediction accuracy. Just two weeks ago we shared a video showcasing $YAI Oracle's training model with an accuracy rate of 0.6703. #Today marks another milestone as we proudly announce a significant leap in predictive precision with a remarkable 23.83% increase. Propelling the Oracle's prediction accuracy to an astounding high of 0.830. 1/

Our development team is breaking new ground in market prediction accuracy. Just two weeks ago we shared a video showcasing $YAI Oracle's training model with an accuracy rate of 0.6703. #Today marks another milestone as we proudly announce a significant leap in predictive precision with a remarkable 23.83% increase. Propelling the Oracle's prediction accuracy to an astounding high of 0.830. 1/

Ÿ

90,429 次观看 • 2 年前

L-14 days We spent 6 hours in the Dragon simulator and were pleasantly surprised to receive our boarding pass at the end of the day. Tomorrow, we’ll have an opportunity to observe Crew-9’s deorbit from Hawthorne and ask questions as it happens. Welcome home, Suni and Butch!

L-14 days We spent 6 hours in the Dragon simulator and were pleasantly surprised to receive our boarding pass at the end of the day. Tomorrow, we’ll have an opportunity to observe Crew-9’s deorbit from Hawthorne and ask questions as it happens. Welcome home, Suni and Butch!

Chun

95,419 次观看 • 1 年前

American shows the price to rent a new movie on Amazon Prime has now risen all the way up to $19.99 (she shows proof) This isn’t to buy it, this is just to rent it for 24 hours It used to cost $3 at Blockbuster and you got to keep it for 3 days

American shows the price to rent a new movie on Amazon Prime has now risen all the way up to $19.99 (she shows proof) This isn’t to buy it, this is just to rent it for 24 hours It used to cost $3 at Blockbuster and you got to keep it for 3 days

Wall Street Apes

577,745 次观看 • 5 个月前

Watch till end 👇👇 1. *LOOSE SKIN GONE*: 3 exercises. 14 days. That’s it. 2. *Tighten + tone your arms without the gym* 3. *Your arms will thank you in 2 weeks* 4. *3 moves to fix underarm "bingo wings"* 5. *Save this: Daily 14-day arm ..👍👍

Watch till end 👇👇 1. LOOSE SKIN GONE: 3 exercises. 14 days. That’s it. 2. Tighten + tone your arms without the gym 3. Your arms will thank you in 2 weeks 4. 3 moves to fix underarm "bingo wings" 5. *Save this: Daily 14-day arm ..👍👍

Skin Care & Nutrition Tips🍸

11,453 次观看 • 18 天前

Full day 3 of our Springball lift for or Line Group. thread. Day 1 and 2 are Squat and Bench Focused. We do cut the volume down by about 1/3 during springball. Priority goes to the lift we get the most MEANINGFUL results from. Low box clean 3x2 at 75%

Full day 3 of our Springball lift for or Line Group. thread. Day 1 and 2 are Squat and Bench Focused. We do cut the volume down by about 1/3 during springball. Priority goes to the lift we get the most MEANINGFUL results from. Low box clean 3x2 at 75%

Conor Hughes

49,170 次观看 • 1 年前

We’ve only had access to GAIA-1 for a few weeks and are discovering new capabilities every day. The results are phenomenal! GAIA isn't just a generative video model, it is a world model, ie. controllable by video, text & action prompts Why is this huge for self driving? Thread:

We’ve only had access to GAIA-1 for a few weeks and are discovering new capabilities every day. The results are phenomenal! GAIA isn't just a generative video model, it is a world model, ie. controllable by video, text & action prompts Why is this huge for self driving? Thread:

Alex Kendall

185,456 次观看 • 3 年前

How about Weevils Baseball taking sole possession of first place in the GAC after a doubleheader sweep of NWOSU? Arkansas-Monticello trailed in both games at one point (7-2 in Game 1 and 10-4 in Game 2), but this offense didn't quit to get two wins by the score of 17-8 and 15-13 respectively. Aidan Garrett homered 4 times on the day, 3 in Game 1 and 1 in Game 2 with 8 total RBI. Chaz Poppy homered 3 times on the day, 2 in Game 1 and 1 in Game 2 with 7 total RBI. Camden Tanner had an excellent game 2, going 4-5 with 2 homers and 7 driven in (5-for-10 in all) #D2Baseball

How about Weevils Baseball taking sole possession of first place in the GAC after a doubleheader sweep of NWOSU? Arkansas-Monticello trailed in both games at one point (7-2 in Game 1 and 10-4 in Game 2), but this offense didn't quit to get two wins by the score of 17-8 and 15-13 respectively. Aidan Garrett homered 4 times on the day, 3 in Game 1 and 1 in Game 2 with 8 total RBI. Chaz Poppy homered 3 times on the day, 2 in Game 1 and 1 in Game 2 with 7 total RBI. Camden Tanner had an excellent game 2, going 4-5 with 2 homers and 7 driven in (5-for-10 in all) #D2Baseball

D2 Baseball

16,676 次观看 • 2 年前

1/ Gemini 2.5 is here, and it’s our most intelligent AI model ever. Our first 2.5 model, Gemini 2.5 Pro Experimental is a state-of-the-art thinking model, leading in a wide range of benchmarks – with impressive improvements in enhanced reasoning and coding and now #1 on Arena by a significant margin. With a model this intelligent, we wanted to get it to people as quickly as possible. Find it on Google AI Studio and in the Google Gemini for Gemini Advanced users now – and in Vertex in the coming weeks. This is the start of a new era of thinking models – and we can’t wait to see where things go from here.

1/ Gemini 2.5 is here, and it’s our most intelligent AI model ever. Our first 2.5 model, Gemini 2.5 Pro Experimental is a state-of-the-art thinking model, leading in a wide range of benchmarks – with impressive improvements in enhanced reasoning and coding and now #1 on Arena by a significant margin. With a model this intelligent, we wanted to get it to people as quickly as possible. Find it on Google AI Studio and in the Google Gemini for Gemini Advanced users now – and in Vertex in the coming weeks. This is the start of a new era of thinking models – and we can’t wait to see where things go from here.

Sundar Pichai

864,455 次观看 • 1 年前

🗣️ New clay incoming! Work started last night on taking off the current clay to make room for the new. We can’t wait to have it all in place and ready for great racing during our #SpringFling event at the end of the month.

🗣️ New clay incoming! Work started last night on taking off the current clay to make room for the new. We can’t wait to have it all in place and ready for great racing during our #SpringFling event at the end of the month.

Volusia Speedway Park

22,783 次观看 • 1 年前

Are you ready for it? 👀 The countdown to #ExperienceAptos begins! Check out the stacked lineup of speakers on Day 1. Together, we’re making Aptos history at our inaugural global festival. Less than 48 hours to go. Tap in 🧵🌅

Are you ready for it? 👀 The countdown to #ExperienceAptos begins! Check out the stacked lineup of speakers on Day 1. Together, we’re making Aptos history at our inaugural global festival. Less than 48 hours to go. Tap in 🧵🌅

Aptos

35,353 次观看 • 1 年前