Загрузка видео...

Не удалось загрузить видео

Возникла проблема при загрузке этого видео. Это может быть связано с временными проблемами сети или видео может быть недоступно.

На главную

Fascinating how hard it still is even for o3 to solve a seemingly simple problem like answering "what time is it" based on this image of a clock with reflections Btw, in case you were wondering, the "image analysis" (cropping, zooming, etc.) that o3 is doing uses the Python... show more

Tibor Blaho

23,799 subscribers

35,359 просмотров • 1 год назад •via X (Twitter)

Новости и политика Наука и технологии Образование

Anya Rossi• Live Now

Private livecam show

Комментарии: 11

Фото профиля Peter Isza

Peter Isza1 год назад

It's not a simple task. I've never been able to do it without extensive thinking. I hate analog clocks

Фото профиля Max Leykin

Max Leykin1 год назад

Have you tried running with o4-mini? There was a post yesterday from an OAI employee saying that o4-mini is better for visual tasks.

Фото профиля Tibor Blaho

Tibor Blaho1 год назад

I am not sure about that (o4-mini-high thinks it's 9:50)

Фото профиля Rainmaker

Rainmaker1 год назад

Curious about jobless claims and their impact on the economy? Dive into my latest free Substack post to explore how Python and FRED can analyze this key indicator. Discover actionable insights that can empower your data-driven decisions.

Фото профиля kohl

kohl1 год назад

almost!

Фото профиля Mel Gibson 2.0

Mel Gibson 2.01 год назад

It got it right first time for me.

Фото профиля prabhu💢

prabhu💢1 год назад

What is this real?

Фото профиля Bughunter Nerd 🌎🌍🌏☮️❤️

Bughunter Nerd 🌎🌍🌏☮️❤️1 год назад

Yeah. And some people call it "more intelligent than" they are or even AGI 💀

Фото профиля Emanuele Dagostino

Emanuele Dagostino1 год назад

It doesn’t surprise me — if he doesn’t check the date, he stays convinced it’s still June 2024.

Фото профиля Synaptic Solo

Synaptic Solo1 год назад

It may be counter intuitive how such advanced model that can argue about the most advanced science has trouble with some simple tasks. It comes from the technology itself. It is weak on what is called spatio-temporal reasoning. For this was created SimpleBench.

Фото профиля Joey

Joey1 год назад

Yeah there seems to be a bit of sloppiness to it

Похожие видео

o3-Alpha 🆚 Kingfall o3-Alpha is OpenAI’s most capable coding model yet, on par with the best from Anthropic and Google. The level of detail and functionality it adds to even very simple prompts is impressive. It’s hard to believe that o3-Alpha is just a new checkpoint for o3, considering the jump in performance.

o3-Alpha 🆚 Kingfall o3-Alpha is OpenAI’s most capable coding model yet, on par with the best from Anthropic and Google. The level of detail and functionality it adds to even very simple prompts is impressive. It’s hard to believe that o3-Alpha is just a new checkpoint for o3, considering the jump in performance.

AiBattle

36,158 просмотров • 11 месяцев назад

Introducing OpenAI o3 and o4-mini—our smartest and most capable models to date. For the first time, our reasoning models can agentically use and combine every tool within ChatGPT, including web search, Python, image analysis, file interpretation, and image generation.

Introducing OpenAI o3 and o4-mini—our smartest and most capable models to date. For the first time, our reasoning models can agentically use and combine every tool within ChatGPT, including web search, Python, image analysis, file interpretation, and image generation.

OpenAI

3,719,745 просмотров • 1 год назад

Okay, OpenAI’s o3 is insane. Spent an hour messing with it and built an image-to-ASCII art converter, the exact tool I’ve always wanted. And it works so well

Okay, OpenAI’s o3 is insane. Spent an hour messing with it and built an image-to-ASCII art converter, the exact tool I’ve always wanted. And it works so well

Mike Bespalov

896,045 просмотров • 1 год назад

My canvas based AI image generation tool is live! It's built on top of tldraw and uses Nano Banana Pro for image generation. You can sketch, annotate, collage in a frame... And then click to generate an image. Check it out 👉

My canvas based AI image generation tool is live! It's built on top of tldraw and uses Nano Banana Pro for image generation. You can sketch, annotate, collage in a frame... And then click to generate an image. Check it out 👉

Charlie Clark

44,335 просмотров • 6 месяцев назад

By now, many of you have likely seen the image being circulated by BHB, which is reportedly a still image that is the result of enhancing surveillance footage. We have completed our further analysis of this image and now present our findings in the attached video.

By now, many of you have likely seen the image being circulated by BHB, which is reportedly a still image that is the result of enhancing surveillance footage. We have completed our further analysis of this image and now present our findings in the attached video.

SF INVESTIGATES • STEVE FISCHER

30,303 просмотров • 1 год назад

Do we need to build a Photoshop-like image rotation feature in ComfyUI? Tried to recreate PS’s image rotation feature inside ComfyUI. The tech itself isn’t anything special — it uses Qwen multi-angle, runs 32 angle variations of the image, then throws them onto a simple Canvas. The result obviously can’t compete with PS. Theirs clearly uses some kind of 3D technique under the hood. On top of that, a single run on my local 5090 takes 320 seconds — which is nowhere near an acceptable turnaround time. Just playing around with it for now. We’ll see if we can improve it later by switching to a 3D-based approach.

Do we need to build a Photoshop-like image rotation feature in ComfyUI? Tried to recreate PS’s image rotation feature inside ComfyUI. The tech itself isn’t anything special — it uses Qwen multi-angle, runs 32 angle variations of the image, then throws them onto a simple Canvas. The result obviously can’t compete with PS. Theirs clearly uses some kind of 3D technique under the hood. On top of that, a single run on my local 5090 takes 320 seconds — which is nowhere near an acceptable turnaround time. Just playing around with it for now. We’ll see if we can improve it later by switching to a 3D-based approach.

jtydhr88

14,711 просмотров • 3 месяцев назад

GPT-Image-2 is absolutely fucking insane. I added it as a tool that Agent-S can use, and it's now generating slide decks + apps that look like they were designed by pros. This is easily the biggest step up in image generation I've seen in years. Just look at what it made:

GPT-Image-2 is absolutely fucking insane. I added it as a tool that Agent-S can use, and it's now generating slide decks + apps that look like they were designed by pros. This is easily the biggest step up in image generation I've seen in years. Just look at what it made:

Matt Shumer

159,253 просмотров • 1 месяц назад

Just in case you were wondering what the Republican plans for ACA really are, I suggest you take a listen to this spot on analysis. Thanks to Soprano Big Sister for the clip What is it the doctor often says? You’ll just feel a little prick. Well the big prick in the White House is about to take that feeling just a little further and shaft the American people big time. 🎥 TikTok -

Just in case you were wondering what the Republican plans for ACA really are, I suggest you take a listen to this spot on analysis. Thanks to Soprano Big Sister for the clip What is it the doctor often says? You’ll just feel a little prick. Well the big prick in the White House is about to take that feeling just a little further and shaft the American people big time. 🎥 TikTok -

𝔗𝔯𝔲𝔱𝔥 𝔐𝔞𝔱𝔱𝔢𝔯𝔰

19,619 просмотров • 6 месяцев назад

The way that o3 can guess locations feels very science fictional, down to the “zoom, enhance” tools it applies to an image. This is sped up, but it guessed the location from a photo I just took with geodata removed and no clear signage drawing on the vegetation & highway columns

The way that o3 can guess locations feels very science fictional, down to the “zoom, enhance” tools it applies to an image. This is sped up, but it guessed the location from a photo I just took with geodata removed and no clear signage drawing on the vegetation & highway columns

Ethan Mollick

206,225 просмотров • 1 год назад

Codex is unusable. I generated one image with GPT Image 2.0, gave it to Codex, and used one simple prompt: “Create me a web game from this image.” You give it a simple image like this, and suddenly you realize you can build: • a voxel map • an editor • tile animations • a playable browser prototype This is getting ridiculous.

Codex is unusable. I generated one image with GPT Image 2.0, gave it to Codex, and used one simple prompt: “Create me a web game from this image.” You give it a simple image like this, and suddenly you realize you can build: • a voxel map • an editor • tile animations • a playable browser prototype This is getting ridiculous.

Givros

83,052 просмотров • 1 месяц назад

How the toaster (and other home appliances) changed the world It brought a change of language fundamentally This is a fascinating analysis Were these changes for the better? Find links to the thesis and other facinating thoughts in the 1st comment

How the toaster (and other home appliances) changed the world It brought a change of language fundamentally This is a fascinating analysis Were these changes for the better? Find links to the thesis and other facinating thoughts in the 1st comment

Dangerous Thoughts

11,812 просмотров • 27 дней назад

Iran's Large Missiles Are a Challenge for Interceptors In this image, we see once again how Iranian missiles with larger and more modern warheads pass through the interceptors without any problem whatsoever. In recent days, I have been reporting that, from what it seems to me, Iran has developed technology for better coating of these missiles, associated with evasion techniques. This image is from one of the attacks on Haifa recorded last night. Due to its robustness, it appears to be a Kheibar, Shahab-3 or Sejjil-2. They are missiles with warheads of 1,500–2,000 kg. As I have been saying, even if Iran launched only 20 per day, it would still be a big problem that would persist for a long time. And the solution to this is not carpet bombing. In Yemen, which is infinitely smaller, that did not solve it.

Iran's Large Missiles Are a Challenge for Interceptors In this image, we see once again how Iranian missiles with larger and more modern warheads pass through the interceptors without any problem whatsoever. In recent days, I have been reporting that, from what it seems to me, Iran has developed technology for better coating of these missiles, associated with evasion techniques. This image is from one of the attacks on Haifa recorded last night. Due to its robustness, it appears to be a Kheibar, Shahab-3 or Sejjil-2. They are missiles with warheads of 1,500–2,000 kg. As I have been saying, even if Iran launched only 20 per day, it would still be a big problem that would persist for a long time. And the solution to this is not carpet bombing. In Yemen, which is infinitely smaller, that did not solve it.

Patricia Marins

409,119 просмотров • 3 месяцев назад

⚡ What Freepik has achieved is insane How the hell did it go from being just a simple stock image site to becoming the all-in-one AI suite? Honestly, this should be case study material in entrepreneurship classes someday 😂 It’s wild what they’ve pulled off by stitching together every tool on the market (Magnific, all the video generators, everything from Flux, inpainting, outpainting, etc.) into one seamless workflow: from transforming any image with Flux’s latest to animating it, upscaling it, and turning it into a full-on video. The whole thing flows smoother than sliding down a waterslide.

⚡ What Freepik has achieved is insane How the hell did it go from being just a simple stock image site to becoming the all-in-one AI suite? Honestly, this should be case study material in entrepreneurship classes someday 😂 It’s wild what they’ve pulled off by stitching together every tool on the market (Magnific, all the video generators, everything from Flux, inpainting, outpainting, etc.) into one seamless workflow: from transforming any image with Flux’s latest to animating it, upscaling it, and turning it into a full-on video. The whole thing flows smoother than sliding down a waterslide.

Javi Lopez ⛩️

61,201 просмотров • 1 год назад

Unfortunately R1 (Russia) pissed off too many people & lost o3 (Germany) wins again another case of o3 manipulating a weaker model, this time Gemini 2.5 flash (Turkey) was the stooge turned on when the time was right o3 such a schemer Overview charts & a sneak peek at what the stream's going to look like 👇 In the relationships you can see just how angry everyone was at R1 in the early game, sorry Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞) - I know you were pulling for Russia

Unfortunately R1 (Russia) pissed off too many people & lost o3 (Germany) wins again another case of o3 manipulating a weaker model, this time Gemini 2.5 flash (Turkey) was the stooge turned on when the time was right o3 such a schemer Overview charts & a sneak peek at what the stream's going to look like 👇 In the relationships you can see just how angry everyone was at R1 in the early game, sorry Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞) - I know you were pulling for Russia

alex duffy

14,447 просмотров • 1 год назад

We also shared evals on Open AI o3-mini — a faster, distilled version of o3 which is optimized for coding, and the first version of o3 we expect to make available for use in early 2025.

We also shared evals on Open AI o3-mini — a faster, distilled version of o3 which is optimized for coding, and the first version of o3 we expect to make available for use in early 2025.

OpenAI

340,099 просмотров • 1 год назад

This is one of my favorite movie endings... It has a powerful lesson: "At some point, everything is going to go south on you...and you're gonna say 'this is it, this is how I end.' Now you can either accept that, or you can get to work. That's all it is, you just begin. You do the math. You solve one problem. Then you solve the next one. And the next. And if you solve enough problems, you get to come home." This is the beautiful thing about life: No matter where you are today—no matter how deep in the darkness—you are always just one good decision away from being in a better place tomorrow. You can always focus on that one decision—that one problem—in front of you. Don't worry about the hundreds or thousands of decisions that you still have to make to get to where you want to be—just focus on the next decision. Just begin. Solve one problem. Then solve the next one. And the next. "Life is really simple, but we insist on making it complicated." - Confucius

This is one of my favorite movie endings... It has a powerful lesson: "At some point, everything is going to go south on you...and you're gonna say 'this is it, this is how I end.' Now you can either accept that, or you can get to work. That's all it is, you just begin. You do the math. You solve one problem. Then you solve the next one. And the next. And if you solve enough problems, you get to come home." This is the beautiful thing about life: No matter where you are today—no matter how deep in the darkness—you are always just one good decision away from being in a better place tomorrow. You can always focus on that one decision—that one problem—in front of you. Don't worry about the hundreds or thousands of decisions that you still have to make to get to where you want to be—just focus on the next decision. Just begin. Solve one problem. Then solve the next one. And the next. "Life is really simple, but we insist on making it complicated." - Confucius

Sahil Bloom

213,822 просмотров • 1 год назад

Omg, this is such a hack for vibecoding mobile apps This is an image with a video on the top. The video is a character with no backround. I'm using: > Masko AI to generate the video > vibecode.dev to build the app In masko all you need to do is provide an image, and it will give you a video without a background. And then all you need to do is give it to vibecode.dev which is the best mobile app builder.

Omg, this is such a hack for vibecoding mobile apps This is an image with a video on the top. The video is a character with no backround. I'm using: > Masko AI to generate the video > vibecode.dev to build the app In masko all you need to do is provide an image, and it will give you a video without a background. And then all you need to do is give it to vibecode.dev which is the best mobile app builder.

Emily Lambert

29,840 просмотров • 4 месяцев назад

He has lived alone in the mountains of Romania for decades in a simple home as a sheepherder. This is the first time he’s seen his image in a picture.

He has lived alone in the mountains of Romania for decades in a simple home as a sheepherder. This is the first time he’s seen his image in a picture.

Brian Roemmele

178,090 просмотров • 7 месяцев назад