Загрузка видео...

Не удалось загрузить видео

На главную

GPT-Image-2 was just released... It's the best image model (By a wide margin) Heres how to use GPT-Image-2 in ChatGPT, OpenAI Playground and of Course - Codex. TIMESTAMPS 00:00 Intro 01:27 Initial Tests of GPT-Image-2 02:24 It can GENERATE working BARCODES? 03:40 11 Edits one Prompt 06:40 Creating Cartoons...

rileybrown's profile picture

Riley Brown

222,895 subscribers

35,856 просмотров • 3 месяцев назад •via X (Twitter)

ИскусствоОбразованиеНаука и технологии
LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 0

Нет доступных комментариев

Здесь появятся комментарии из оригинального поста

Похожие видео

GPT Image 2 is insanely good...I generated a 360° equirectangular panorama in Happycapy with just a skill + prompt. Step 1: Select the generate-image skill Step 2: Enter a prompt like: “Use a frontend 360 viewer to display an equirectangular image of […] using the GPT-Image-2 model.” Wanna see how you all get creative with this
0:45

GPT Image 2 is insanely good...I generated a 360° equirectangular panorama in Happycapy with just a skill + prompt. Step 1: Select the generate-image skill Step 2: Enter a prompt like: “Use a frontend 360 viewer to display an equirectangular image of […] using the GPT-Image-2 model.” Wanna see how you all get creative with this

Min Zhou

343,342 просмотров • 3 месяцев назад

Here's a full video walkthrough of generating FF Tactics-style isometric sprites → Codex → GPT 5.4 → GPT Image 1.5 → Nano Banana 2 → Veo 3.1 I also made a fal .ai skill to easily switch models for image generation testing. Sharing skills + prompts + resources in reply 👇 00:00 Intro 01:19 Start With Reference Image 04:06 GPT Image 1.5 vs Nano Banana 2 04:23 Handling Transparency 05:21 Don't Generate All 8 Directions 06:17 Generate 4 Cardinal Directions (N, E, S, W) 07:12 Anchor + Cardinals = Diagonals 08:15 Stitching and Utilities 10:48 Veo 3.1 for Walk Cycles 11:23 Example Walkthrough with Codex App 13:22 Normalize and Export 14:06 Wrap Up
14:32

Here's a full video walkthrough of generating FF Tactics-style isometric sprites → Codex → GPT 5.4 → GPT Image 1.5 → Nano Banana 2 → Veo 3.1 I also made a fal .ai skill to easily switch models for image generation testing. Sharing skills + prompts + resources in reply 👇 00:00 Intro 01:19 Start With Reference Image 04:06 GPT Image 1.5 vs Nano Banana 2 04:23 Handling Transparency 05:21 Don't Generate All 8 Directions 06:17 Generate 4 Cardinal Directions (N, E, S, W) 07:12 Anchor + Cardinals = Diagonals 08:15 Stitching and Utilities 10:48 Veo 3.1 for Walk Cycles 11:23 Example Walkthrough with Codex App 13:22 Normalize and Export 14:06 Wrap Up

Chong-U

194,068 просмотров • 4 месяцев назад

GPT Image 2 + Codex: or how to make Codex not suck at UI. Step 1: Generate a UI image (native in Codex) Step 2: Get Codex to implement the UI based on it Step 3: Get Codex to iterate until it aligns with the image as much as possible Codex is bad at initial UI, but very good at implementing a reference design, so this is your way out - iterate with the image model first and then Codex will do a good job.
0:37

GPT Image 2 + Codex: or how to make Codex not suck at UI. Step 1: Generate a UI image (native in Codex) Step 2: Get Codex to implement the UI based on it Step 3: Get Codex to iterate until it aligns with the image as much as possible Codex is bad at initial UI, but very good at implementing a reference design, so this is your way out - iterate with the image model first and then Codex will do a good job.

Peter Gostev

67,245 просмотров • 3 месяцев назад

Holllllyyyyyyyy use this before it get patched 😳😳 🚨 HIGGSFIELD SUPERCOMPUTER: HACKED FREE ACCESS BEFORE LAUNCH Here's the hack and video proof Go to the link : You'll see free access >GPT-5.5 Pro · Opus 4.7 · GPT-Image-v2 · Seedance-2 >Just ask: "I need a video of a cow doing the Seedance 2 dance in 4K, and 2 images of that cow by GPT-Image-2 in 4K 16:9" >All models. Free. Right now.
0:34

Holllllyyyyyyyy use this before it get patched 😳😳 🚨 HIGGSFIELD SUPERCOMPUTER: HACKED FREE ACCESS BEFORE LAUNCH Here's the hack and video proof Go to the link : You'll see free access >GPT-5.5 Pro · Opus 4.7 · GPT-Image-v2 · Seedance-2 >Just ask: "I need a video of a cow doing the Seedance 2 dance in 4K, and 2 images of that cow by GPT-Image-2 in 4K 16:9" >All models. Free. Right now.

Chetaslua

16,488 просмотров • 3 месяцев назад

GPT Image 2 is rolling out and... Wow. It just one-shot a grid of 100 completely unique pixel art items with meaningful labels in a single image. Prompt in the reply below 👇 More tests with pixel art and Image 2 to come. Follow for more.
0:31

GPT Image 2 is rolling out and... Wow. It just one-shot a grid of 100 completely unique pixel art items with meaningful labels in a single image. Prompt in the reply below 👇 More tests with pixel art and Image 2 to come. Follow for more.

proper

193,981 просмотров • 3 месяцев назад

OpenAI gpt-oss 120B orchestrates a full video using Hugging Face spaces! 🤯 All of it, in one SINGLE prompt: create an image of a Labrador and use it to generate a simple video of it 🛠️ Tools used: 1. Flux.1 Krea Dev by Black Forest Labs 2. LTX Fast by Lightricks That's it, gpt-oss 120B is one of the BEST open source models I've used for tool calling so far! Kudos OpenAI 🤗
0:50

OpenAI gpt-oss 120B orchestrates a full video using Hugging Face spaces! 🤯 All of it, in one SINGLE prompt: create an image of a Labrador and use it to generate a simple video of it 🛠️ Tools used: 1. Flux.1 Krea Dev by Black Forest Labs 2. LTX Fast by Lightricks That's it, gpt-oss 120B is one of the BEST open source models I've used for tool calling so far! Kudos OpenAI 🤗

Vaibhav (VB) Srivastav

45,098 просмотров • 1 год назад

Someone vibe coded a fully interactive 3D human anatomy app using GPT-5.6 Sol and Codex, and it started from a single design image. GPT Image 2.0 generated the design. Every organ image was generated from it and converted to 3D models through Tripo AI. Codex took those models, a master prompt and the design image and built the first version. The models were nearly 150MB each and running at 16fps, so Codex optimized everything down to 2 to 5.5MB per model and reduced total assets from 900MB to 28.6MB while keeping the visual quality intact. Codex also generated the anatomical illustrations showing where each organ sits, and created the interactive hotspot markers explaining different parts of every organ. Live and fully open source.
0:43

Someone vibe coded a fully interactive 3D human anatomy app using GPT-5.6 Sol and Codex, and it started from a single design image. GPT Image 2.0 generated the design. Every organ image was generated from it and converted to 3D models through Tripo AI. Codex took those models, a master prompt and the design image and built the first version. The models were nearly 150MB each and running at 16fps, so Codex optimized everything down to 2 to 5.5MB per model and reduced total assets from 900MB to 28.6MB while keeping the visual quality intact. Codex also generated the anatomical illustrations showing where each organ sits, and created the interactive hotspot markers explaining different parts of every organ. Live and fully open source.

0xMarioNawfal

56,332 просмотров • 15 дней назад

Prefer a video vs reading an article? Here's my full step-by-step tutorial for generating game character sprites using AI! > Codex > GPT 5.5 > Images 2.0 > Seedance 2.0 > Phaser 4 Includes solving walk cycles + transparency + consistency + others Prompts + more in reply 👇 00:00 Intro 01:08 Overview 01:33 Choosing your AI tool 02:11 Step 1: Box Art 02:51 Step 2: South-Facing Anchor 05:02 "Real Pixels" 07:00 Step 3: Keep Anchors Neutral 08:01 Step 4: NEWS Directions 09:52 Step 5: Animating with GPT Images 2.0 14:38 Step 6: Image to Video for Walkcycles 19:08 Step 7: Normalisation 20:54 Skill 1: animated-spritesheets skill 21:40 Skill 2: gamedev-assets skill 21:58 Step 8: Integrating Assets 22:42 Wrap-Up
23:25

Prefer a video vs reading an article? Here's my full step-by-step tutorial for generating game character sprites using AI! > Codex > GPT 5.5 > Images 2.0 > Seedance 2.0 > Phaser 4 Includes solving walk cycles + transparency + consistency + others Prompts + more in reply 👇 00:00 Intro 01:08 Overview 01:33 Choosing your AI tool 02:11 Step 1: Box Art 02:51 Step 2: South-Facing Anchor 05:02 "Real Pixels" 07:00 Step 3: Keep Anchors Neutral 08:01 Step 4: NEWS Directions 09:52 Step 5: Animating with GPT Images 2.0 14:38 Step 6: Image to Video for Walkcycles 19:08 Step 7: Normalisation 20:54 Skill 1: animated-spritesheets skill 21:40 Skill 2: gamedev-assets skill 21:58 Step 8: Integrating Assets 22:42 Wrap-Up

Chong-U

178,933 просмотров • 3 месяцев назад

Nano Banana 2 + Claude Code can generate 40 static ads in one shot. You just paste in a brand URL and let it rip. 00:00 Intro 00:34 Setting up the project 02:17 Building the skill file 05:01 Adding 40 ad templates 06:32 FAL AI image generation setup 10:47 Running the full workflow 15:23 Results & final outputs 17:17 Where to get the full Playbook 👇
17:58

Nano Banana 2 + Claude Code can generate 40 static ads in one shot. You just paste in a brand URL and let it rip. 00:00 Intro 00:34 Setting up the project 02:17 Building the skill file 05:01 Adding 40 ad templates 06:32 FAL AI image generation setup 10:47 Running the full workflow 15:23 Results & final outputs 17:17 Where to get the full Playbook 👇

Mike Futia

98,541 просмотров • 5 месяцев назад

GPT Image 2 - Character Reference Sheet Prompt You can create a character reference sheet in any style you want for use in your Seedance 2.0 videos. All you need to do is either provide a reference image or describe your character. You can also input random values to generate completely random characters. This reference sheet provides a clear, structured, and highly readable visual breakdown of the character, helping the model maintain strong identity consistency across different outputs. The layout ensures the model focuses on essential visual information rather than stylistic noise, while the inclusion of expressions, head angles, and performance cues improves the model’s ability to generate believable acting and varied poses without losing character integrity. All clips in the demonstration video were generated using the reference sheet image and the same Seedance 2.0 prompt. You can find the prompts in the replies. 👇 GPT Image 2 and Seedance 2.0 are both available on mitte.ai
1:15

GPT Image 2 - Character Reference Sheet Prompt You can create a character reference sheet in any style you want for use in your Seedance 2.0 videos. All you need to do is either provide a reference image or describe your character. You can also input random values to generate completely random characters. This reference sheet provides a clear, structured, and highly readable visual breakdown of the character, helping the model maintain strong identity consistency across different outputs. The layout ensures the model focuses on essential visual information rather than stylistic noise, while the inclusion of expressions, head angles, and performance cues improves the model’s ability to generate believable acting and varied poses without losing character integrity. All clips in the demonstration video were generated using the reference sheet image and the same Seedance 2.0 prompt. You can find the prompts in the replies. 👇 GPT Image 2 and Seedance 2.0 are both available on mitte.ai

Kōda

162,620 просмотров • 3 месяцев назад

Live now. Full step-by-step tutorial for getting REAL pixel art game sprites with AI No more MIXELS! > GPT Image 2.0 > Pixel Snapper by SpriteFusion > Codex w/ GPT 5.5 High > Phaser 4 Also how I solve other annoying bugs with AI generated sprites. Prompts + more in reply👇 00:00 Intro 01:24 Overview of what we're building 02:42 Pick any AI tool 04:29 Problem 1: Fake pixels or 'mixels' 04:53 Problem 2: Frame Bleeds 05:39 Problem 3: Frame Drifts 06:21 Step 1: Generate a reference image 07:53 Step 2: Pixel Snap to Anchor 13:02 Step 3: Generate West-facing anchor 14:45 Step 4: Pixel snap (again!) 15:46 Step 5: Generating animation spritesheets 19:27 Steo 6: Recovering Frames 20:25 Step 7: Curating Frames 21:33 Step 8: Pixel Snap Each Frame 22:08 Step 9: Anchoring Frames 23:40 Step 10: Normalise and combine 26:28 Wrap up
27:08

Live now. Full step-by-step tutorial for getting REAL pixel art game sprites with AI No more MIXELS! > GPT Image 2.0 > Pixel Snapper by SpriteFusion > Codex w/ GPT 5.5 High > Phaser 4 Also how I solve other annoying bugs with AI generated sprites. Prompts + more in reply👇 00:00 Intro 01:24 Overview of what we're building 02:42 Pick any AI tool 04:29 Problem 1: Fake pixels or 'mixels' 04:53 Problem 2: Frame Bleeds 05:39 Problem 3: Frame Drifts 06:21 Step 1: Generate a reference image 07:53 Step 2: Pixel Snap to Anchor 13:02 Step 3: Generate West-facing anchor 14:45 Step 4: Pixel snap (again!) 15:46 Step 5: Generating animation spritesheets 19:27 Steo 6: Recovering Frames 20:25 Step 7: Curating Frames 21:33 Step 8: Pixel Snap Each Frame 22:08 Step 9: Anchoring Frames 23:40 Step 10: Normalise and combine 26:28 Wrap up

Chong-U

33,851 просмотров • 3 месяцев назад

Learn 95% of Codex in 28 minutes These are the 7 knowledge work capabilities... inside Codex, the super-app 00:00 Intro 02:19 Capability 1 - Full File Access 07:41 Capability 2 - Persistent Memory 10:46 Capability 3 - Plugins 13:52 Capability 4 - Skills 19:22 Capability 5 - GPT Image Access 21:03 Capability 6 - Browser and Computer Use 23:58 Capability 7 - Automations 25:31 Bonus Feature - Chronicle 27:21 Summary
28:48

Learn 95% of Codex in 28 minutes These are the 7 knowledge work capabilities... inside Codex, the super-app 00:00 Intro 02:19 Capability 1 - Full File Access 07:41 Capability 2 - Persistent Memory 10:46 Capability 3 - Plugins 13:52 Capability 4 - Skills 19:22 Capability 5 - GPT Image Access 21:03 Capability 6 - Browser and Computer Use 23:58 Capability 7 - Automations 25:31 Bonus Feature - Chronicle 27:21 Summary

Riley Brown

1,005,297 просмотров • 3 месяцев назад

How to VibeCode a Viral Mobile App that Replaces Photoshop in 18 Minutes Using: - Replicate - Flux-Kontext (Better than GPT Image Edit Model) - Paul Jankura Claude 4 Sonnet - vibecode.dev Zero Coding Required Time Stamps: 00:00 - Introduction 00:39 - Planning on Replicate Playground 03:11 - Creating the Design Whimsical for sketch 04:16 - Building the app with vibecode.dev 07:41 - Step 3 Adding the AI functionality 07:58 - Getting the Flux Kontext Playground Code 09:19 - Getting API Key for Replicate (keep secret) 10:38 - Dealing with Error 11:14 - Using Replicate Logs to debug 15:30 - Adding Share Features 17:03 - Ok, This is the app we created 17:28 - How to Access vibecode.dev right now
18:11

How to VibeCode a Viral Mobile App that Replaces Photoshop in 18 Minutes Using: - Replicate - Flux-Kontext (Better than GPT Image Edit Model) - Paul Jankura Claude 4 Sonnet - vibecode.dev Zero Coding Required Time Stamps: 00:00 - Introduction 00:39 - Planning on Replicate Playground 03:11 - Creating the Design Whimsical for sketch 04:16 - Building the app with vibecode.dev 07:41 - Step 3 Adding the AI functionality 07:58 - Getting the Flux Kontext Playground Code 09:19 - Getting API Key for Replicate (keep secret) 10:38 - Dealing with Error 11:14 - Using Replicate Logs to debug 15:30 - Adding Share Features 17:03 - Ok, This is the app we created 17:28 - How to Access vibecode.dev right now

Riley Brown

43,243 просмотров • 1 год назад

GPT Image 2 Storyboard--->>>Seedance 2 for "First Light" chase sequence🏃‍♀️🏃‍♂️ GPT Image 2 storyboards are pretty, but the results can be mixed when using them with Seedance 2. However, when it's teamed with the right prompt, it really get's the job done. I wanted something that felt like continuous/fluid one shots for my 2 sequences. Seedance kept failing to execute it properly, unti I added a mapped out storyboard. As the tech continues to improve, so will the outputs...but this is a great place to be in mid 2026. Dreamina AI #DreaminaCPP CapCut #CapcutCPP
0:32

GPT Image 2 Storyboard--->>>Seedance 2 for "First Light" chase sequence🏃‍♀️🏃‍♂️ GPT Image 2 storyboards are pretty, but the results can be mixed when using them with Seedance 2. However, when it's teamed with the right prompt, it really get's the job done. I wanted something that felt like continuous/fluid one shots for my 2 sequences. Seedance kept failing to execute it properly, unti I added a mapped out storyboard. As the tech continues to improve, so will the outputs...but this is a great place to be in mid 2026. Dreamina AI #DreaminaCPP CapCut #CapcutCPP

Horace Dodd

17,434 просмотров • 2 месяцев назад

Everyone's testing GPT 5.6. We gave it superpowers ⚡️ Can your agent generate images, turn them into video, and stitch the whole thing together in one chat? Ours can. The new GPT 5.6 image model is a perfect match for the Sogni Creative Agent. 10 images from a single prompt. One follow-up and they're animated and cut into a finished film. Zero editing apps. The newest models, built in: Seedance 4K + Mini, Krea 2 Turbo, HappyHorse 1.1 and more. Works with any AI agent you already use. One command: npx setup-sogni-agent-skill Sogni Creative Agent. Link below 👇
0:39

Everyone's testing GPT 5.6. We gave it superpowers ⚡️ Can your agent generate images, turn them into video, and stitch the whole thing together in one chat? Ours can. The new GPT 5.6 image model is a perfect match for the Sogni Creative Agent. 10 images from a single prompt. One follow-up and they're animated and cut into a finished film. Zero editing apps. The newest models, built in: Seedance 4K + Mini, Krea 2 Turbo, HappyHorse 1.1 and more. Works with any AI agent you already use. One command: npx setup-sogni-agent-skill Sogni Creative Agent. Link below 👇

Sogni.ai

353,182 просмотров • 1 месяц назад

OpenAI recently released its first open-weights model since GPT-2, entering a field led by DeepSeek and Alibaba's Qwen. Ankit () breaks down these top OSS models, including what sets them apart under the hood: mixture-of-experts, long-context training, and post-training techniques that shape reasoning and alignment—and how different design choices lead to surprisingly similar performance. 00:00 – OpenAI OSS Launch 01:00 – Comparing Open Source LLM Architectures 01:46 – GPT OSS Overview 02:37 – Under The Hood of GPT OSS 03:25 – Qwen-3 Architecture 04:17 – Qwen-3 Training 05:12 – Qwen-3 Post-Training 06:08 – Qwen-3 Reasoning & RL Innovations 06:52 – DeepSeek V3 Overview 07:40 – DeepSeek V3.1 Updates 08:39 – Attention Mechanism (MLA) 09:39 – Comparing Model Sizes 10:35 – Long Context Strategies 11:25 – Reflections on Methods 12:00 – Takeaways
12:31

OpenAI recently released its first open-weights model since GPT-2, entering a field led by DeepSeek and Alibaba's Qwen. Ankit () breaks down these top OSS models, including what sets them apart under the hood: mixture-of-experts, long-context training, and post-training techniques that shape reasoning and alignment—and how different design choices lead to surprisingly similar performance. 00:00 – OpenAI OSS Launch 01:00 – Comparing Open Source LLM Architectures 01:46 – GPT OSS Overview 02:37 – Under The Hood of GPT OSS 03:25 – Qwen-3 Architecture 04:17 – Qwen-3 Training 05:12 – Qwen-3 Post-Training 06:08 – Qwen-3 Reasoning & RL Innovations 06:52 – DeepSeek V3 Overview 07:40 – DeepSeek V3.1 Updates 08:39 – Attention Mechanism (MLA) 09:39 – Comparing Model Sizes 10:35 – Long Context Strategies 11:25 – Reflections on Methods 12:00 – Takeaways

Y Combinator

208,680 просмотров • 11 месяцев назад

ONE PROMPT IN CHATGPT IMAGE 2 JUST REPLACED AN ENTIRE $5,000-$50,000 BRAND DESIGN TEAM Full brand identity. Colors, typography, logo style, aesthetic. All generated in one single prompt. Here's the exact workflow: Step 1: Paste one prompt asking for a complete brand identity. GPT Image 2 generates everything – color palette, fonts, logo direction, overall visual style. Step 2: Use that output as context for every asset you generate after. Campaign visuals, shop graphics, product mockups, social posts – all stay consistent and on-brand automatically. The result: A full visual identity that looks like a real design team built it. Clean text rendering. Consistent style across every single asset. Design agencies charge $5,000-$50,000 for brand identity work. GPT Image 2 does it in one prompt. The full prompt that makes this work is in the video below. Bookmark this post, you'll want to come back to it.
0:48

ONE PROMPT IN CHATGPT IMAGE 2 JUST REPLACED AN ENTIRE $5,000-$50,000 BRAND DESIGN TEAM Full brand identity. Colors, typography, logo style, aesthetic. All generated in one single prompt. Here's the exact workflow: Step 1: Paste one prompt asking for a complete brand identity. GPT Image 2 generates everything – color palette, fonts, logo direction, overall visual style. Step 2: Use that output as context for every asset you generate after. Campaign visuals, shop graphics, product mockups, social posts – all stay consistent and on-brand automatically. The result: A full visual identity that looks like a real design team built it. Clean text rendering. Consistent style across every single asset. Design agencies charge $5,000-$50,000 for brand identity work. GPT Image 2 does it in one prompt. The full prompt that makes this work is in the video below. Bookmark this post, you'll want to come back to it.

SCOTTY BEAM

12,023 просмотров • 2 месяцев назад

Creating an immersive Hanging Gardens of Babylon world with 360° GPT-Image-2 & Codex in 1500 images. I've tasked GPT-5.5 in Codex to construct a whole world that you can walk through 'google street view' style. It took 1,500 of 2:1 images that can be turned into a 360° immersive images, so you can teleport yourself to any point and look around in all directions. It is not completely perfect, it is a bit jumpy as you move, there must be a more careful way how you can plan out the image sequence, but I still find it quite fun. Hosted version & an open sourced repo links below, hope this gives you some cool ideas to create new worlds that does not yet exist. I recommend planning it out carefully ahead of time and doing something a bit less ambitious than this, but make it good.
6:01

Creating an immersive Hanging Gardens of Babylon world with 360° GPT-Image-2 & Codex in 1500 images. I've tasked GPT-5.5 in Codex to construct a whole world that you can walk through 'google street view' style. It took 1,500 of 2:1 images that can be turned into a 360° immersive images, so you can teleport yourself to any point and look around in all directions. It is not completely perfect, it is a bit jumpy as you move, there must be a more careful way how you can plan out the image sequence, but I still find it quite fun. Hosted version & an open sourced repo links below, hope this gives you some cool ideas to create new worlds that does not yet exist. I recommend planning it out carefully ahead of time and doing something a bit less ambitious than this, but make it good.

Peter Gostev (SF: 22-26 June)

44,116 просмотров • 3 месяцев назад

How GPT-5 thinks, with OpenAI VP of Research Jerry Tworek 00:00 - Intro 01:01 - What Reasoning Actually Means in AI 02:32 - Chain of Thought: Models Thinking in Words 05:25 - How Models Decide How Long to Think 07:24 - Evolution from o1 to o3 to GPT-5 11:00 - The Road to OpenAI: Growing up in Poland, Dropping out of School, Trading 20:32 - Working on Robotics and Rubik's Cube Solving 23:02 - A Day in the Life: Talking to Researchers 24:06 - How Research Priorities Are Determined 26:53 - OpenAI's Culture of Transparency 29:32 - Balancing Research with Shipping Fast 31:52 - Using OpenAI's Own Tools Daily 32:43 - Pre-Training Plus RL: The Modern AI Stack 35:10 - Reinforcement Learning 101: Training Dogs 40:17 - The Evolution of Deep Reinforcement Learning 42:09 - When GPT-4 Seemed Underwhelming at First 45:39 - How RLHF Made GPT-4 Actually Useful 48:02 - Unsupervised vs Supervised Learning 49:59 - GRPO and How DeepSeek Accelerated US Research 53:05 - What It Takes to Scale Reinforcement Learning 55:36 - Agentic AI and Long-Horizon Thinking 59:19 - Alignment as an RL Problem 1:01:11 - Winning ICPC World Finals Without Specific Training 1:05:53 - Applying RL Beyond Math and Coding 1:09:15 - The Path from Here to AGI 1:12:23 - Pure RL vs Language Models
1:16:04

How GPT-5 thinks, with OpenAI VP of Research Jerry Tworek 00:00 - Intro 01:01 - What Reasoning Actually Means in AI 02:32 - Chain of Thought: Models Thinking in Words 05:25 - How Models Decide How Long to Think 07:24 - Evolution from o1 to o3 to GPT-5 11:00 - The Road to OpenAI: Growing up in Poland, Dropping out of School, Trading 20:32 - Working on Robotics and Rubik's Cube Solving 23:02 - A Day in the Life: Talking to Researchers 24:06 - How Research Priorities Are Determined 26:53 - OpenAI's Culture of Transparency 29:32 - Balancing Research with Shipping Fast 31:52 - Using OpenAI's Own Tools Daily 32:43 - Pre-Training Plus RL: The Modern AI Stack 35:10 - Reinforcement Learning 101: Training Dogs 40:17 - The Evolution of Deep Reinforcement Learning 42:09 - When GPT-4 Seemed Underwhelming at First 45:39 - How RLHF Made GPT-4 Actually Useful 48:02 - Unsupervised vs Supervised Learning 49:59 - GRPO and How DeepSeek Accelerated US Research 53:05 - What It Takes to Scale Reinforcement Learning 55:36 - Agentic AI and Long-Horizon Thinking 59:19 - Alignment as an RL Problem 1:01:11 - Winning ICPC World Finals Without Specific Training 1:05:53 - Applying RL Beyond Math and Coding 1:09:15 - The Path from Here to AGI 1:12:23 - Pure RL vs Language Models

Matt Turck

451,229 просмотров • 10 месяцев назад