Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

GPT-5.5 by Reasoning Effort: I've asked it in Codex to create a physics-based visualisation of RL cycles for different sized models (70b, 1t, 10t), to demonstrate how the amount of RL you can do differs by model size. My assessment of each: - Low: weird slop - Medium: kinda...

petergostev's profile picture

Peter Gostev

11,591 subscribers

208,732 Aufrufe • vor 1 Monat •via X (Twitter)

BildungWissenschaft & Technologie
LIVE

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

We released physics-intern: a simple harness for science problems! It gets models like Gemini 3.1 Pro to go from 17.7 -> 31.4, thus beating GPT 5.5 Pro. The physics-intern harness can wrap any model and via dedicated subagent boost the performance of the vanilla reasoning models. While I think more and more of these harness capability gains will be absorbed into the models (like prompting tricks disappeared over time) there is a lot to be gained right now by building good scaffolds for those models and integrating tools well. Interestingly, the exception we found that GPT 5.5 Pro actually didn't benefit from the physics-intern harness! Read more about it here: PS: I think the Harness[Model] notation is kind of nice.
0:12

We released physics-intern: a simple harness for science problems! It gets models like Gemini 3.1 Pro to go from 17.7 -> 31.4, thus beating GPT 5.5 Pro. The physics-intern harness can wrap any model and via dedicated subagent boost the performance of the vanilla reasoning models. While I think more and more of these harness capability gains will be absorbed into the models (like prompting tricks disappeared over time) there is a lot to be gained right now by building good scaffolds for those models and integrating tools well. Interestingly, the exception we found that GPT 5.5 Pro actually didn't benefit from the physics-intern harness! Read more about it here: PS: I think the Harness[Model] notation is kind of nice.

Leandro von Werra

96,161 Aufrufe • vor 25 Tagen

Asked GPT-5.5 in Codex to reconstruct this model from the image. Pretty close!
0:12

Asked GPT-5.5 in Codex to reconstruct this model from the image. Pretty close!

Ruben Kostandyan

124,735 Aufrufe • vor 1 Monat

BREAKING: Anthropic just dropped Opus 4.8—and it is a MONSTER We've been testing for about a week Every 📧 and our verdict is they could've just called it Opus 5, it's that good. Here's our vibe check: - Beats GPT-5.5 on Senior Engineer bench. On our toughest benchmark Opus 4.8 scores a 63—a hair higher than GPT-5.5's score of 62, and a full 30 points higher than Opus 4.7. It tackled a ground-up rewrite of a production codebase, and actually built something that works. HOWEVER: Coding performance varied a lot at different reasoning levels. We recommend using it on xhigh for best results. - Incredibly good writer. Opus 4.8 scored a 79.6 on our writing benchmark—measuring models on real-world writing tasks we do all of the time like essay writing, promo email writing, and more. It beats GPT-5.5 by 6 points. It produces well-written prose with fewer "AI-isms". It's also very good at writing in your voice given the right context. HOWEVER: Writing performance also varied with reasoning levels. Medium reasoning had higher incidence of AI-isms—we found best results with high. - Beast at knowledge work. Opus 4.8 is very good at general knowledge work tasks like report creation, research and more. It produced the best PowerPoint one-shot we've ever seen on our deck generation benchmark. - Emotionally intelligent, willing to question the frame. I've also found it to be quite good at talking through psychological or interpersonal issues. It has a high EQ, and it's also good at not glazing and helping to expand your perspective. Its thought process feels extremely rich and dynamic. THE BAD: These days a model is only as good as its harness, and Codex is still a far superior harness to the Claude Desktop app. This has kept me using Codex + GPT-5.5 as my daily driver, but I am flipping back and forth a lot more between Codex and Claude. Anthropic is back baby! Read the rest on Every 📧:
0:11

BREAKING: Anthropic just dropped Opus 4.8—and it is a MONSTER We've been testing for about a week Every 📧 and our verdict is they could've just called it Opus 5, it's that good. Here's our vibe check: - Beats GPT-5.5 on Senior Engineer bench. On our toughest benchmark Opus 4.8 scores a 63—a hair higher than GPT-5.5's score of 62, and a full 30 points higher than Opus 4.7. It tackled a ground-up rewrite of a production codebase, and actually built something that works. HOWEVER: Coding performance varied a lot at different reasoning levels. We recommend using it on xhigh for best results. - Incredibly good writer. Opus 4.8 scored a 79.6 on our writing benchmark—measuring models on real-world writing tasks we do all of the time like essay writing, promo email writing, and more. It beats GPT-5.5 by 6 points. It produces well-written prose with fewer "AI-isms". It's also very good at writing in your voice given the right context. HOWEVER: Writing performance also varied with reasoning levels. Medium reasoning had higher incidence of AI-isms—we found best results with high. - Beast at knowledge work. Opus 4.8 is very good at general knowledge work tasks like report creation, research and more. It produced the best PowerPoint one-shot we've ever seen on our deck generation benchmark. - Emotionally intelligent, willing to question the frame. I've also found it to be quite good at talking through psychological or interpersonal issues. It has a high EQ, and it's also good at not glazing and helping to expand your perspective. Its thought process feels extremely rich and dynamic. THE BAD: These days a model is only as good as its harness, and Codex is still a far superior harness to the Claude Desktop app. This has kept me using Codex + GPT-5.5 as my daily driver, but I am flipping back and forth a lot more between Codex and Claude. Anthropic is back baby! Read the rest on Every 📧:

Dan Shipper 📧

350,456 Aufrufe • vor 18 Tagen

it's crazy what a 1.5B model can do these days! "VibeThinker-1.5B is a 1.5-billion parameter dense language model. With a total training cost of only $7,800 USD, it achieves reasoning performance comparable to larger models like GPT OSS-20B Medium." runs perfectly on device!
0:18

it's crazy what a 1.5B model can do these days! "VibeThinker-1.5B is a 1.5-billion parameter dense language model. With a total training cost of only $7,800 USD, it achieves reasoning performance comparable to larger models like GPT OSS-20B Medium." runs perfectly on device!

Maziyar PANAHI

202,273 Aufrufe • vor 7 Monaten

My favorite paper this year: "Video models are zero-shot learners and reasoners" It illustrates that video models show emergent visual reasoning at scale - they can solve vision tasks they weren't trained for. This may be the "GPT moment" for vision. Let's break it down 👇
0:19

My favorite paper this year: "Video models are zero-shot learners and reasoners" It illustrates that video models show emergent visual reasoning at scale - they can solve vision tasks they weren't trained for. This may be the "GPT moment" for vision. Let's break it down 👇

Justine Moore

185,001 Aufrufe • vor 5 Monaten

GPT-5.5 + GPT Image 2 in Codex is insane 🤯 The watch components generation + alignment was one shot.
0:11

GPT-5.5 + GPT Image 2 in Codex is insane 🤯 The watch components generation + alignment was one shot.

Katia Gil Guzman

759,739 Aufrufe • vor 1 Monat

Opus 4.6 vs GPT 5.4 (High) (1/9) prompt: Build a single-file HTML/CSS/JS (no libs) demo that uses SVG to simulate a plant growing: stem extends, leaves sprout + unfurl with springy/windy “physics”, then seamlessly loops forever. For the initial impressions I'm really impressed by GPT, for speed they both felt about the same, but gpt was still half cheaper than opus. I also much prefer the design and animation that gpt produced, physics on the leaves are super cool and it also loops pretty nicely whilst opus just fades out the plant. Still got a bunch of tests to run but this is really exciting.
0:10

Opus 4.6 vs GPT 5.4 (High) (1/9) prompt: Build a single-file HTML/CSS/JS (no libs) demo that uses SVG to simulate a plant growing: stem extends, leaves sprout + unfurl with springy/windy “physics”, then seamlessly loops forever. For the initial impressions I'm really impressed by GPT, for speed they both felt about the same, but gpt was still half cheaper than opus. I also much prefer the design and animation that gpt produced, physics on the leaves are super cool and it also loops pretty nicely whilst opus just fades out the plant. Still got a bunch of tests to run but this is really exciting.

Dev Ed

659,944 Aufrufe • vor 3 Monaten

This is the game GPT-5.5 built for me in one shot, featured on the release page. What makes it different from previous AI-generated games is how fun it actually feels to play. The controls are smooth, and every shape is generated from scratch in Three.js.
0:12

This is the game GPT-5.5 built for me in one shot, featured on the release page. What makes it different from previous AI-generated games is how fun it actually feels to play. The controls are smooth, and every shape is generated from scratch in Three.js.

Pietro Schirano

80,965 Aufrufe • vor 1 Monat

You can now quickly eval GPT-5 and reasoning efforts across your existing responses. With the built-in grader, compare responses to find the best model and reasoning effort for you. ⚡️👀
0:13

You can now quickly eval GPT-5 and reasoning efforts across your existing responses. With the built-in grader, compare responses to find the best model and reasoning effort for you. ⚡️👀

OpenAI Developers

79,899 Aufrufe • vor 10 Monaten

Excited to introduce OpenAI o1—a new series of reasoning models. Developers, we’d love for you to kick the tires, but don’t just hot-swap GPT-4o with o1! Send o1-preview and o1-mini your coding tasks or hard challenges, and share your feedback and most interesting results.
0:15

Excited to introduce OpenAI o1—a new series of reasoning models. Developers, we’d love for you to kick the tires, but don’t just hot-swap GPT-4o with o1! Send o1-preview and o1-mini your coding tasks or hard challenges, and share your feedback and most interesting results.

Romain Huet

234,323 Aufrufe • vor 1 Jahr

🚀 New Video Alert – GPT-5 in Copilot Studio! GPT-5 has officially arrived in Microsoft Copilot Studio, and I couldn’t resist putting it to the test. In this video, I explore: 🔍 GPT-5 Auto & GPT-5 Reasoning models for custom agents. 🛠 How I leveraged GPT-5 Chat inside a custom prompt for more tailored outputs. ⚖️ GPT-5 Chat Vs. GPT-4.1 to highlight improvements and trade-offs If you’ve been wondering what GPT-5 brings to Copilot Studio—this video is packed with real examples, practical tests, and insights to help you understand what these new experimental models have to offer. 🎥 Watch here: #CopilotStudio #GPT5 #AIBuilder #MicrosoftCopilot #LowCodeAI #OpenAI #AI #CopilotStudio #GPT5Auto #GPT5Reasoning #GPT5Chat
0:16

🚀 New Video Alert – GPT-5 in Copilot Studio! GPT-5 has officially arrived in Microsoft Copilot Studio, and I couldn’t resist putting it to the test. In this video, I explore: 🔍 GPT-5 Auto & GPT-5 Reasoning models for custom agents. 🛠 How I leveraged GPT-5 Chat inside a custom prompt for more tailored outputs. ⚖️ GPT-5 Chat Vs. GPT-4.1 to highlight improvements and trade-offs If you’ve been wondering what GPT-5 brings to Copilot Studio—this video is packed with real examples, practical tests, and insights to help you understand what these new experimental models have to offer. 🎥 Watch here: #CopilotStudio #GPT5 #AIBuilder #MicrosoftCopilot #LowCodeAI #OpenAI #AI #CopilotStudio #GPT5Auto #GPT5Reasoning #GPT5Chat

Daniel Christian

14,705 Aufrufe • vor 10 Monaten

Rejoice. Just following up with another quick W in Codex You can now configure your reasoning level in plan mode separately directly from your config file. This is huge for Plus users who want to plan with high or xhigh reasoning levels, and then switch over to medium reasoning for implementation, without needing the slash command. This is a great way to save your usage limits, and now it happens automatically. Even if you're on Pro, this should make you very happy. Prior to this, it was switching you automatically to medium every time you planned, which was pretty annoying. Place this near the top of your config file: plan_mode_reasoning = "high" (or xhigh) 0.150.0 is a massive quality of life update. They're clearly listening. This time I am shouting out Charlie. 🙏
0:17

Rejoice. Just following up with another quick W in Codex You can now configure your reasoning level in plan mode separately directly from your config file. This is huge for Plus users who want to plan with high or xhigh reasoning levels, and then switch over to medium reasoning for implementation, without needing the slash command. This is a great way to save your usage limits, and now it happens automatically. Even if you're on Pro, this should make you very happy. Prior to this, it was switching you automatically to medium every time you planned, which was pretty annoying. Place this near the top of your config file: plan_mode_reasoning = "high" (or xhigh) 0.150.0 is a massive quality of life update. They're clearly listening. This time I am shouting out Charlie. 🙏

am.will

25,712 Aufrufe • vor 3 Monaten

Kimi K2.5 vs GPT-5.2 codex same prompt , different results - svg of birds < i am using Yupp to compare these models , like i dont need to buy subscription for each companies to compare lol 🫣>
0:19

Kimi K2.5 vs GPT-5.2 codex same prompt , different results - svg of birds < i am using Yupp to compare these models , like i dont need to buy subscription for each companies to compare lol 🫣>

Chetaslua

14,926 Aufrufe • vor 4 Monaten

I've been doing the F-Zero test each time a new model comes out and GPT 5.5 and Codex is the only combo that built a working game so far. Even made a bunch of other bots to race against. What a insane time to be building.
0:13

I've been doing the F-Zero test each time a new model comes out and GPT 5.5 and Codex is the only combo that built a working game so far. Even made a bunch of other bots to race against. What a insane time to be building.

Peter Yang

149,186 Aufrufe • vor 1 Monat

GPT Image 2 + Seedance 2 is an insane combo the fact that you can get really high detail and animate it is so satisfying people will kill me for this take, but this is the future of game development
0:10

GPT Image 2 + Seedance 2 is an insane combo the fact that you can get really high detail and animate it is so satisfying people will kill me for this take, but this is the future of game development

INK

116,379 Aufrufe • vor 1 Monat

Just tried Baidu’s new ERNIE X1.1 and wow… this is actually impressive 🤯 I’ve been testing it on complex coding problems and math reasoning tasks, and the jump from earlier models is night and day. What stood out to me: ▪️ Much stronger at factual accuracy and instruction following ▪️ Handles agent-like tasks, tool use, and multi-step reasoning smoothly ▪️ Far fewer hallucinations compared to other reasoning-heavy models ▪️ Solid performance across creative writing, Q&A, coding, and math Benchmarks put ERNIE X1.1 ahead of DeepSeek R1-0528 and on par with GPT-5 & Gemini 2.5 Pro. If you’re into AI or reasoning models, it's definitely worth checking out: This doesn’t feel like just another release — it feels like a new standard for reasoning AI.
0:11

Just tried Baidu’s new ERNIE X1.1 and wow… this is actually impressive 🤯 I’ve been testing it on complex coding problems and math reasoning tasks, and the jump from earlier models is night and day. What stood out to me: ▪️ Much stronger at factual accuracy and instruction following ▪️ Handles agent-like tasks, tool use, and multi-step reasoning smoothly ▪️ Far fewer hallucinations compared to other reasoning-heavy models ▪️ Solid performance across creative writing, Q&A, coding, and math Benchmarks put ERNIE X1.1 ahead of DeepSeek R1-0528 and on par with GPT-5 & Gemini 2.5 Pro. If you’re into AI or reasoning models, it's definitely worth checking out: This doesn’t feel like just another release — it feels like a new standard for reasoning AI.

Parul Gautam

68,406 Aufrufe • vor 9 Monaten

Zero-shot video reasoning (chain-of-frames) isn’t just for Veo3 — open-source models can understand and edit too! 🕹️ ChronoEdit brings temporal reasoning to image editing. 🔗
0:13

Zero-shot video reasoning (chain-of-frames) isn’t just for Veo3 — open-source models can understand and edit too! 🕹️ ChronoEdit brings temporal reasoning to image editing. 🔗

Xuanchi Ren

22,653 Aufrufe • vor 8 Monaten

Updating gpt-taste skill ⏳ this site was built with GPT 5.4 medium in Codex using gpt-taste prompt was "landing page. 8sections. for an AI agency"
0:14

Updating gpt-taste skill ⏳ this site was built with GPT 5.4 medium in Codex using gpt-taste prompt was "landing page. 8sections. for an AI agency"

Leon Lin

25,714 Aufrufe • vor 1 Monat

The AI voice agent in my Apple Watch is now ready! Model: gpt-realtime-2 Thanks to Codex 5.5 xhigh Coming soon: it will be able to see my calendar, create reminders, take notes, and more tools
0:13

The AI voice agent in my Apple Watch is now ready! Model: gpt-realtime-2 Thanks to Codex 5.5 xhigh Coming soon: it will be able to see my calendar, create reminders, take notes, and more tools

Diego | AI 🚀 - e/acc

22,275 Aufrufe • vor 1 Monat

You can ask Codex to update an existing repo to GPT-5.5.
0:16

You can ask Codex to update an existing repo to GPT-5.5.

OpenAI Developers

185,406 Aufrufe • vor 1 Monat