avv

@thdxr This is daxxing…

im never leaving this app

@athyuttamre A huge improvement over the earlier voice mode. Is access through api expected soon?

GPT-Live is now fully rolled out to all ChatGPT users globally. We're also doubling everyone's voice usage limit for the whole weekend so you can try more of it. Have fun! 🎉

🔥 Thats great to see. Just 1 request - OpenClaw is starting to get complicated and is starting to have a huge barrier for people not already familiar with it. A simpler usage variant (even if it has a lower number of integrations / skills) can go a long way in exposing more people to its possibilities.

Today we’re introducing the OpenClaw Foundation: a nonprofit home for open, independent personal AI. A full-time team. Great partners. One mission: bring personal AI to everyone. Welcome to the age of the lobster.🦞 openclaw.ai/blog/introduci…

@Angaisb_ Already here.. in DE

GPT Live 1 is so cool I hope it also gets released in the EU!

@martian588 Raised a couple of PR's. That said, have some questions about the direction of the project. Possible to DM?

Looking for contributors - Devlaner/devlane: Open-source Jira, Linear, Monday, ClickUp and Plane alternative. github.com/Devlaner/devla…

@tavus would you be supporting the new GPT-Live mode in your PALs over the coming weeks?

@OpenAI Hopefully, Europe too 🙏 🤞

Introducing GPT-Live, a new generation of voice models for natural human-AI interaction. Rolling out in ChatGPT starting today. You’ll want to turn the sound on for this one.

Sure, there is a monetisation guardrail that would exist for ad-funded consumer apps. But for B2B software, it may end up being a stronger default

While it may seem very incremental, I do believe self-generative apps (like Monogram) are likely to be a UX that most companies would make default going forward. Why have one fixed view, when everyone see's exactly what they care most about.

@wafer_ai thats a crazy number, any idea on time to first token (ttft) yet?

🚨 OpenAI is launching GPT-5.6 Sol on Cerebras at up to 750 tps. Here's How kernels work on Cerebras Chips cerebras built a chip with 900,000 cores on a single silicon wafer. CSL (Cerebras Software Language) is a Zig-inspired DSL that gives you direct control over the Wafer-Scale Engine. the SDK is publicly documented and the programming model is genuinely the most alien one we've ever looked at. in CUDA you write one thread's perspective and launch millions. in CSL, there are no threads, no warps, no shared memory, and no kernel launches. you write code for individual Processing Elements: 900,000 independent cores arranged in a 2D mesh on a single silicon wafer. each PE has its own 48 KB of private SRAM, its own program counter, and a 5-port router connecting it to its 4 neighbors. that's actually it. no DRAM. no HBM. no cache hierarchy. 48 KB is your entire world per PE, where code and data must both fit. the programming model is dataflow. data moves between PEs as 32-bit messages called wavelets, traveling along virtual channels called colors. when a wavelet arrives at a PE on a specific color, it activates a task (a chunk of code bound to that color at compile time). tasks run to completion, then hardware picks the next activated task. tasks cannot call each other. they can only be activated. so instead of "launch N threads," you think of it like: "place code on PEs, define routes, let data flow." the memory model is also very different from GPUs. on an H100 you get 80 GB of HBM shared across all SMs. on the WSE-3, memory is 48 KB per PE, and there are 900,000 of them! this gives you 44 GB total on-chip SRAM at 21 PB/s aggregate bandwidth (vs 3 TB/s on H100). every access is single-cycle. no coalescing needed. no bank conflicts. no cache misses. but also no way to access another PE's memory. all inter-PE communication is explicit wavelet routing through the fabric. take a distributed GEMV for example. you would write a layout file that physically routes wavelets across the mesh. two PEs sit side by side. the left PE computes a partial result and routes it eastward. the right PE receives those wavelets from the west and accumulates. routing is defined at compile time. both operations are asynchronous and activate a task when they finish. you're physically routing data across silicon at 1 clock cycle per hop. cerebras is an extremely technically interesting & alien beast. reports 95-210x speedups over H100 on stencil computations. 3,000 tokens/sec inference on gpt-oss-120B. $10B+ deal with OpenAI for 750 MW of inference infrastructure. very exciting times for alternative accelerators! deep dive 1/6 by @gpuemi

@blader Any feedback on the speed when used with the API? Sam had mentioned it does 750 toks / sec, but no mention of TTFT anyplace yet

i've been using gpt 5.6 sol ultra as my workhorse for some time, and what i can say right now is that it's a very, very good model if they launched this as gpt 6 i don't think anybody would really question it more to say on launch day

@thsottiaux Any feedback on the speed when used with the API? @sama had a comment on the 750 toks / sec, but no mention of TTFT anyplace yet (for real time voice turn use cases)

Prepare your sunglasses. Sol is coming. 😎

@sama Any feedback on the speed - specifically TTFT? 750 toks / sec is awesome, but how would it play with low latency use cases @OpenAIDevs ?

GPT-5.6 sol launches thursday! happy building

@skirano Any feedback on the speed when used with the API? @sama had a comment on the 750 toks / sec, but no mention of TTFT anyplace yet (for real time voice turn use cases)

I can finally talk about 5.6. I’ve been testing it for months and, without exaggeration, it’s the best model I’ve ever used. Fast, smart, genuinely creative, and you guessed it, they finally fixed front-end design. I haven’t needed to check the code I’ve written in two months.

@Teknium @theemozilla

@thsottiaux Visual dashboards : having a visual front to a repeatable process within codes, that doesnt require me to create a app, and view in the browser.

What is something that you feel is surprising that Codex still can't do well and we should have gotten right a while ago?

A more intuitive workflow layer: define personas, responsibilities, and handoffs clearly. Instead of managing everything through long prompts, users could create role-based workflows. For example, a designer explores the experience, an architect defines the solution, and a developer builds it. Each persona has a clear responsibility, relationship to the others, and expected output.