Loading video...

Video Failed to Load

Go Home

This is dots3-note preview from dots studio in a puzzle game it hasn't seen before. No game-specific prompt. It has to work out the controls, objective and new mechanics from the screen. It starts pressing buttons. Open-weight: 280B total / 16B active, 512K context.

khushbooverma_'s profile picture

Khushboo Verma

164,547 subscribers

37,624 views • 3 days ago •via X (Twitter)

Science & TechnologyEducationGaming
LIVE

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

rednote's AI lab, dots studio dots studio, just released a new open-weight model: dots3-note Preview → 280B parameters / 16B active → 512K context → Text + vision + speech Built for something most agents still struggle with: tasks that run for hours, days, or even weeks Here’s how:
1:39

rednote's AI lab, dots studio dots studio, just released a new open-weight model: dots3-note Preview → 280B parameters / 16B active → 512K context → Text + vision + speech Built for something most agents still struggle with: tasks that run for hours, days, or even weeks Here’s how:

Nelly;

37,245 views • 2 days ago

Meet dots3-note preview — an open-weight multimodal model built for long-horizon agency in real life. → 280B total / 16B active parameters → 512K context window → Native text, vision, and speech understanding Designed to reason, use tools, self-evaluate, and adapt as tasks evolve over time. dots studio
0:42

Meet dots3-note preview — an open-weight multimodal model built for long-horizon agency in real life. → 280B total / 16B active parameters → 512K context window → Native text, vision, and speech understanding Designed to reason, use tools, self-evaluate, and adapt as tasks evolve over time. dots studio

Alamin

56,808 views • 8 days ago

A road trip plan is never just a simple route. Learn more: Api free now: Weather shifts. Highways close. Strategies must adapt. Real-world challenges demand AI that thrives in uncertainty. This is exactly what dots studio's dots3-note preview is designed to solve. It is an open-weight multimodal model boasting 280B total parameters, 16B active parameters, and a 512K context window. The true differentiator is TEMPO. During long-horizon tasks, agents require more than just memory — they must constantly verify they are still tracking toward the correct objective. Leveraging an Actor-Critic approach, TEMPO critically evaluates choices mid-action, course-correcting strategies long before the task concludes. Combined with: 🌍 Multimodal understanding — tackling real-world problems by processing text, images, and audio as one cohesive input. 🌱 Learning in deployment — adapting to unfamiliar environments through ongoing exploration and active feedback. From mapping a dynamic journey to navigating intricate problems, dots3-note preview marks a new era of AI agency. This exact vision fuels rednote, showcasing raw reasoning capabilities that hit IMO gold medal-level performance.
1:27

A road trip plan is never just a simple route. Learn more: Api free now: Weather shifts. Highways close. Strategies must adapt. Real-world challenges demand AI that thrives in uncertainty. This is exactly what dots studio's dots3-note preview is designed to solve. It is an open-weight multimodal model boasting 280B total parameters, 16B active parameters, and a 512K context window. The true differentiator is TEMPO. During long-horizon tasks, agents require more than just memory — they must constantly verify they are still tracking toward the correct objective. Leveraging an Actor-Critic approach, TEMPO critically evaluates choices mid-action, course-correcting strategies long before the task concludes. Combined with: 🌍 Multimodal understanding — tackling real-world problems by processing text, images, and audio as one cohesive input. 🌱 Learning in deployment — adapting to unfamiliar environments through ongoing exploration and active feedback. From mapping a dynamic journey to navigating intricate problems, dots3-note preview marks a new era of AI agency. This exact vision fuels rednote, showcasing raw reasoning capabilities that hit IMO gold medal-level performance.

SOUL AI

10,365 views • 3 days ago

The objective purpose of Game Design is to make it so EVERY Action game in existence has this much density from the first minute
0:57

The objective purpose of Game Design is to make it so EVERY Action game in existence has this much density from the first minute

Blader

19,443 views • 4 months ago

Plan my New Zealand trip is a prompt. The trip itself is not. Once the task stretches across time, the agent has to deal with accumulated decisions, unfamiliar situations, and changing conditions—not just generate another answer. That’s the problem dots studio’s dots3-note preview is designed to explore.
1:27

Plan my New Zealand trip is a prompt. The trip itself is not. Once the task stretches across time, the agent has to deal with accumulated decisions, unfamiliar situations, and changing conditions—not just generate another answer. That’s the problem dots studio’s dots3-note preview is designed to explore.

Evan Cipher 🧠

25,778 views • 8 days ago

Progress in open models is keeping Big AI labs up at night, and I'm here for it! We have a brand new open-weight multimodal model optimized for long-horizon tasks. This model is really good at something: it can work on tasks that keep evolving over time. • 280B total parameters, but only 16B active • 512K context window • Understands text, images, and audio • Strong reasoning, coding, and tool use But the best of all: the model learns and adapts to new information! Imagine you start running an agent today to solve a problem, and while it's working, you get new information that changes the initial conditions, or you change your mind. The agents you run today don't have issues with short tasks and goals that don't change, but reality is messy, and that makes it hard for long-horizon agents to succeed. The new dots3-note Preview model introduces TEMPO. TEMPO is a new reinforcement learning technique that lets the model periodically pause and critique its own progress. Basically, from time to time, the agent asks itself: "Am I getting closer to the goal, or am I wasting my time?" The same model switches between actor and critic. The actor works on the problem. The critic looks at the current state, reasons about how much progress it has made, and determines what should happen next. TEMPO gives the model feedback along the way. This is huge for any agent that can work on long-horizon tasks without wasting its time.
4:05

Progress in open models is keeping Big AI labs up at night, and I'm here for it! We have a brand new open-weight multimodal model optimized for long-horizon tasks. This model is really good at something: it can work on tasks that keep evolving over time. • 280B total parameters, but only 16B active • 512K context window • Understands text, images, and audio • Strong reasoning, coding, and tool use But the best of all: the model learns and adapts to new information! Imagine you start running an agent today to solve a problem, and while it's working, you get new information that changes the initial conditions, or you change your mind. The agents you run today don't have issues with short tasks and goals that don't change, but reality is messy, and that makes it hard for long-horizon agents to succeed. The new dots3-note Preview model introduces TEMPO. TEMPO is a new reinforcement learning technique that lets the model periodically pause and critique its own progress. Basically, from time to time, the agent asks itself: "Am I getting closer to the goal, or am I wasting my time?" The same model switches between actor and critic. The actor works on the problem. The critic looks at the current state, reasons about how much progress it has made, and determines what should happen next. TEMPO gives the model feedback along the way. This is huge for any agent that can work on long-horizon tasks without wasting its time.

Santiago

79,311 views • 8 days ago

You're not gonna find a more immersive gaming experience with your NFT like this... Own your experience and share it with other Forgotten Runes Wizard's Cult holders. From amazing in-game mechanics to social aspects this game has it all. Check out The Forgotten Runiverse Lets find out what the future of web3 gaming has to offer...
4:53

You're not gonna find a more immersive gaming experience with your NFT like this... Own your experience and share it with other Forgotten Runes Wizard's Cult holders. From amazing in-game mechanics to social aspects this game has it all. Check out The Forgotten Runiverse Lets find out what the future of web3 gaming has to offer...

Unboxing Web3

12,767 views • 1 year ago

Ayo, marketing a game sucks so here is my game. It has morse It has puzzles It has morse puzzles You can walk and do morse puzzles There is a link to a (free!!) demo in my bio. 👇Example of a baller puzzle in the game👇
0:29

Ayo, marketing a game sucks so here is my game. It has morse It has puzzles It has morse puzzles You can walk and do morse puzzles There is a link to a (free!!) demo in my bio. 👇Example of a baller puzzle in the game👇

IronDev

29,163 views • 7 months ago

We are going to release a new beta version of Mumu Tycoon tmr at 9am UTC on Feb 10, 2025. It hasn't added tokenomics in the game yet, but it has included the most completed game with all new features for production. You can start familiar with the game before P2E starts. 🔥🔥🔥
0:20

We are going to release a new beta version of Mumu Tycoon tmr at 9am UTC on Feb 10, 2025. It hasn't added tokenomics in the game yet, but it has included the most completed game with all new features for production. You can start familiar with the game before P2E starts. 🔥🔥🔥

Mumubit

79,128 views • 1 year ago

“Honestly, it is a new game and a new day, so we wipe it, learn from it, and came out here and played a good game today.” 🗣️ Hear more from #SeaKraken forward Berkly Catton following tonight’s big win over the New Jersey Devils.
1:57

“Honestly, it is a new game and a new day, so we wipe it, learn from it, and came out here and played a good game today.” 🗣️ Hear more from #SeaKraken forward Berkly Catton following tonight’s big win over the New Jersey Devils.

Seattle Kraken

12,219 views • 7 months ago

Slots & Daggers is OUT NOW!! I've poured so much love into this game and had so much fun making it. From the music to the animations to all the silly little mechanics, really hope you can feel it 🖤 Give it a retweet if you wanna support me, link to the game below! Cheers 🥂
1:09

Slots & Daggers is OUT NOW!! I've poured so much love into this game and had so much fun making it. From the music to the animations to all the silly little mechanics, really hope you can feel it 🖤 Give it a retweet if you wanna support me, link to the game below! Cheers 🥂

Friedemann 🎰🗡️ Slots & Daggers

122,329 views • 10 months ago

We've forgotten that this game came out 6 years ago on the PS4, yet it still has the best gameplay mechanics of the decade. No current-gen game has topped it... and I honestly don't think anything will for another 6 years.
0:49

We've forgotten that this game came out 6 years ago on the PS4, yet it still has the best gameplay mechanics of the decade. No current-gen game has topped it... and I honestly don't think anything will for another 6 years.

Nyx

42,347 views • 26 days ago

How did Abubakar Salim cast @the_ben_starr and @neilnewbon in Dead Take? The new puzzle game from @SurgentStudios offers a "horrific take" on Hollywood, and it even features an ancient headshot from Neil Newbon's early career. Check out the full interview:
0:54

How did Abubakar Salim cast @the_ben_starr and @neilnewbon in Dead Take? The new puzzle game from @SurgentStudios offers a "horrific take" on Hollywood, and it even features an ancient headshot from Neil Newbon's early career. Check out the full interview:

Radio Times

13,371 views • 1 year ago

Rushing Beat Ran (Jaleco/1993) is a 2D side-scrolling beat 'em up for the Super Nintendo. Fast and fun, it's the second game in Jaleco's Rushing Beat trilogy. In the West, it was heavily localized and stripped of its cutscenes, released under the title Brawl Brothers. But it's possible to unlock the original by pressing the B, A, X, and Y buttons repeatedly during the Jaleco logo screen until it will glitch. Then, press right to move the glitched screen right, then press down 3 times and then Start to access the (glitched) options menu. There, the game starts to work normally as Rushing Beat Ran afterwards.
0:59

Rushing Beat Ran (Jaleco/1993) is a 2D side-scrolling beat 'em up for the Super Nintendo. Fast and fun, it's the second game in Jaleco's Rushing Beat trilogy. In the West, it was heavily localized and stripped of its cutscenes, released under the title Brawl Brothers. But it's possible to unlock the original by pressing the B, A, X, and Y buttons repeatedly during the Jaleco logo screen until it will glitch. Then, press right to move the glitched screen right, then press down 3 times and then Start to access the (glitched) options menu. There, the game starts to work normally as Rushing Beat Ran afterwards.

Dante retro dev 💾

13,676 views • 2 months ago

New Game: Crossfire from the x-devs of Naughty Dogs & Infinity Ward. It is a narrative-driven, third-person, single player, action-adventure game. Story - 2 people from opposite ends join forces together to overcome an obstacle. They must work together to survive Studio: That's No Moon
0:49

New Game: Crossfire from the x-devs of Naughty Dogs & Infinity Ward. It is a narrative-driven, third-person, single player, action-adventure game. Story - 2 people from opposite ends join forces together to overcome an obstacle. They must work together to survive Studio: That's No Moon

SpawnYaard 🎮

21,111 views • 2 months ago

A fun experiment for academics: Paste your favorite paper into Gemini 2.5 (assuming open rights) and ask it to turn it into a working game with "Canvas" turned on. This was in one go, I just trimmed out the waiting times from the video. I love the references to specific tables.
1:44

A fun experiment for academics: Paste your favorite paper into Gemini 2.5 (assuming open rights) and ask it to turn it into a working game with "Canvas" turned on. This was in one go, I just trimmed out the waiting times from the video. I love the references to specific tables.

Ethan Mollick

122,992 views • 1 year ago

Anyone know how to fix this? Game has not worked since this new update, the things I catch from research recognise where I am, and it lets me join remote raids but that’s it, the game won’t work outside of that @NianticHelp Pokémon GO
0:29

Anyone know how to fix this? Game has not worked since this new update, the things I catch from research recognise where I am, and it lets me join remote raids but that’s it, the game won’t work outside of that @NianticHelp Pokémon GO

Jaytheunderdog

12,846 views • 2 years ago