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 cooked - High: sort of tried but ultimately incoherent - Extra High: elite - really nice idea and well executed Obviously this is just one shot, but worth trying different reasoning levels for the new models, medium seems to be pretty good for GPT-5.5 and it was really bad for many previous GPT models.show more

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.show more

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 📧:show more

You can now use GPT 5.5, Gemini 3.7 Flash,... Kimi K3 and 47 other AI models completely free😱 No subscription. No credit card. Even the API usage costs $0. AIHubMix just opened a free catalog with 50 AI models. Some of the available models: • Ox Alpha • Gemini 3.7 Flash • GLM 5.2 • Kimi K3 • MiniMax M3 • GPT 5.5 • 40+ more And you don’t need separate API keys for each model. Setup takes 2 minutes: > Step 1: Go to > Create an account using your email or OAuth. No card needed. Step 2: Create one API key > The same key works with every free and paid model. Step 3: Add it to any OpenAI-compatible tool Base URL: Then choose any model ending in -free, such as: coding-glm-5.2-free gpt-5.5-free That’s it. One API key. 50 AI models. $0 for both input and output. Save this. You might need a free multi-model setup later.show more

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.show more

You can now orchestrate Fable 5, Sol, and any... model inside Codex with one plugin. It's called Codex-Orchestration. Assign Fable 5 as the advisor, Sol as the executor, or any model to any role. Then define the order they work in. Codex handles the routing. I ran Fable 5 High as planner with GPT-5.6 Sol Extra High as executor on a set of issues Opus and GPT-5.5 always struggled with. Done in 30 minutes. 40% fewer limit hits. 2x faster implementation. Install it by pasting this into Codex: "Install Codex Orchestration: codex plugin marketplace add Cjbuilds/Codex-Orchestration codex plugin add codex-orchestration@codex-orchestration Verify the installation, then tell me to start a new task." Then assign your models: @ codex-orchestration advisor: Claude Fable 5 High, Executor: GPT-5.6 Sol High Open source. Tweak the routing however you want.show more

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. 🙏show more

Claude Opus 4.7 vs GPT 5.4 (2/7) Testing how... well these models do with cloth physics simulation, and honestly I'm quite disappointed to see how poorly GPT 5.4 performed here. Performance was pretty bad... it added all the functionality (panning, cutting, dragging the curtain) under one mouse event, causing it to glitch out a lot. The scene also looks kinda shit. Opus 4.7 pretty much nailed the prompt. First of all, the angle and scene it picked are really nice, with the lamp just creeping into the frame. The dragging, wind physics, and cutting all worked. I also had to reprompt gpt 1-2 times to properly build and remove a bunch of ugly ui cards it added on top of the 3d scene. prompt: Build a premium Vite + React interactive cloth simulation of a cozy room curtain using real particle/constraint physics, with structural + shear springs, damping, collisions, and stable real-time deformation. Users should be able to drag the curtain, add wind, and cut constraints, while the experience stays polished, visually elegant, and numerically stable.show more

for those that don't grasp how crazy this is... GPT Image 2.5 Sunburst vs Ideogram 4.5 and by the way it's likely just simple masking/in-painting but it still surprises me it's not the default for the majority of image editing models. you can tell it's in-painting because you can see the bush around my shoulder slowly turn into a line. it's the same reason my tattoos fade on why side of my arm (the closest to my t-shirt). no excuse for chopping of half of my head though. I'm going to have to try a workflow where I use this on every frame of a video.show more

The redesigned Arena Leaderboard Overview is live! The frontier... is moving quickly, so we built a central hub to help you keep up. The new Arena Leaderboard experience brings all our live signals into one place for researchers, builders, and our global community of users who power our model evaluations. Here’s what’s new: - New Release Models: a real-time feed of the latest models added to Arena - Performance by Category: surfacing live sessions, and the top models by categories like Agents, Images, and Coding. - Model Capabilities: first impressions of new models, straight from the Arena team - Arena News: the latest research, features and updates from around the Arena universe For example, now you can see scores across arenas for models like GPT-6 Sol and Claude Opus 5.5, alongside signals like their category standings and capabilities, their latest news and more. Explore scores and now much more at our redesigned leaderboard. Find the link below.show more

ChatGPT Web is now inside Codex 😲 this open-source... project has already crossed 2.7k stars instead of using a separate workflow, it lets you use ChatGPT Web models directly from Codex's model picker what you get: - GPT-5.6 Pro for eligible accounts - free Luna access - ChatGPT Web quota - Codex tools + context - images, streaming and reasoning - open-source + MIT licensed getting started: 1. go to 2. install the launcher 3. sign in with your ChatGPT account 4. run the browser checks 5. install the models 6. restart Codex and select ChatGPT Web the interesting part? you can keep using Codex normally while routing the selected model through ChatGPT Web no separate API key for the ChatGPT model 2.7k+ stars and still actively updated worth checking if you already use Codex and want to experiment with ChatGPT Web modelsshow more

One of the craziest use cases I’ve found for... Jev: verifiers. I am so excited about this that I at least wanted to share the high-level idea. I used Jev to build a custom verifier for the /goal feature in my agent harness. It checks whether the goal is actually complete after every turn, making continuous verification cheap enough to scale. This means I can run more of these verifiers (previously handled by another expensive reasoning model) more frequently to keep the agents on track. System One models are perfect for verification. I think of this as scaling harnesses further by cleverly combining System One and System Two models. I have a feeling this will enable a new wave of scalable test-time compute methods. Watch this space closely. I've just started to experiment with this and am already seeing really good results. I need to explore and figure out a way to benchmark it. I will share more once I have more results. This is an insane unlock for long-horizon agents. You heard it here first. And you can expect to see more harnesses embracing this new pattern. Full guide dropping in the next couple of days.show more

🚨 Fable 5.5 XHigh vs GPT 6.1 Sol Ultra... Fable 5.5 now seems to be routing through both the web app and Claude Code for most of us, On my Claude Pro x5 plan, this test used around 7% of my weekly limit without any agents, Running the same kind of test on Codex x20 used only about 1% of my weekly limit Quality wise, both were really strong. In terms of speed, they were also pretty close, with both taking roughly 30 minutes to finish, I’m running a lot more tests in parallel right now to see whether there’s actually a meaningful difference between them once you push both models hardershow more

Fable 5 totally crushed our new contest, but it... cost 6x more than Opus 4.8! We gave 4 models the same prompt: build three self-contained HTML5 canvas scenes with real physics demos Prompts: — A train derailing off a broken bridge into the water — Two cars jumping off ramps and colliding mid-air over a canyon — A monster truck crushing a row of parked cars Outputs: Fable 5: 62,158 tokens, $3.12 GPT 5.5: 37,753 tokens, $1.14 Opus 4.8: 22,280 tokens, $0.56 GLM 5.2: 36,246 tokens, $0.08 Fable 5 did all three scenes at A+. The crashes looked real, things fell and broke the right way, and nothing went through the ground or floated. GPT 5.5 was the closest to Fable. In the Bigfoot show, we think GPT was even a little better. GLM 5.2 did not win any scene, but it was the cheapest by far. Fable is the best pick for quality, but you pay more for it.show more

GPT 6.1 Sol Ultra vs DeepSeek V4.1 Flash 1017... Max, I still can't believe a model that's 90% cheaper can produce results this good, how is this even possible? Both models took roughly the same amount of time since most of it was spent rendering, but the quality DeepSeek managed to achieve is incredible, especially considering the API cost was just 0.44 cents for the entire test! I think having access to free web search is making a huge difference and giving DeepSeek a much bigger advantage than before, especially with its native multimodal capabilities. The more I test this model, the more I wonder how they're managing to offer so much at such a low price, I'm still running even bigger tests with different models to see how far it can go, more results coming soon!show more

Alrighty, everything is ready 😎 here’s an unofficial “2x... Codex limits” promo from my side for you all. meet DevSpace — an MCP connector app that turns ChatGPT into Codex. npm install -g @waishnav/devspace After installing, tunnel the MCP server over the internet and enjoy 2x limits. You can use GPT-5.5 Pro, xHigh, or High for planning, then hand off the task to your local Codex/pi/opencode/cursor/claude code instance. Or you can just use it for reviewing code written by other local coding agents Go ahead, experiment with different workflows, and keep the feedback coming on GitHub Issues or in my DMs And let’s thank OpenAI for being so generous by giving us separate ChatGPT and Codex limits and by being so chill around this MCP :) Please use it sparingly, only when you run out of limits. Don’t overuse it — in the end, they do have a button to stop it 🙂show more

🚁 Adding an Apache attack helicopter today I think... I will rewrite the entire flight model today again to be ambiguous of what kind of vehicle it is Planes behave very different than helicopters and different than vehicles like tanks or pedestrians Because people here also requested "Guy with bazooka" so that will be next I have no idea if everyone else is just lying about using AI for their plane models but mine look so clunky (but I like the look), but I now tried with Claude 3.5 and 3.7 and Grok 3 and they all just generate clunky models, not beautiful like you see in other AI vibe coded games, so either they bs'ing or I'm retarded I also have to move sub parts of each model like for 30 minutes until everything fits, AI has a really hard time with 3d models But okay the look is kinda cuteshow more

I wanted to test one of the models out... and so I made a chica walking cycle for the fun of it. ============= Please note that it's not suppose to be really accurate to the movie, but have that taste from the movie. THIS IS ALSO NOT RELATED TO MY PROJECT!!show more

I’m fascinated by GPT-6 Sol and the new ways... it expresses itself through visual elements. And yes, for creative writing, it’s the best so far in my opinion. I’m testing a whole range of use cases, including 18+ stuff, slightly naughtier scenarios, more serious situations, basically trying to see how it reacts across every kind of mature conversation and in different areas. (Don’t mind the content in the video, it was just a test chat for the graphical elements.)show more

This is now the sixth publicly shown interception of... a Russian Geran-5 by a Ukrainian drone interceptor. The interceptor remains unnamed, but judging by its interface it appears to be different from the three models shown previously. If so, this would make it the fourth interceptor type publicly demonstrated against Geran-5. This would mean the 1020th Regiment now operates at least two different high-speed interceptor models: STING S and this new system. Unlike previous interceptions, engagement details were disclosed: the Geran-5 was engaged at 425 km/h and an altitude of over 3 km.show more