Taymour Khan

I built Corso Labs: ED Documentation Review Room for the Gemma 4 hackathon. It uses Gemma 4 31B on @Cerebras to run an 8-agent, multimodal ED documentation QA workflow over a synthetic raster medical-record image. The demo reads facts from pixels, runs a visible agent chain, returns a clinician-editable physician note review packet, and compares Cerebras speed against a GPU path. The point: speed changes the product shape. On slow inference, multi-agent review feels batch-like. On Cerebras, it can feel interactive. Live app: corso-labs-review-room.vercel.app Code: github.com/stormyeyez/cor… Synthetic data only. Not diagnosis, treatment, clinical validation, or production software. @googlegemma @GoogleDeepMind @cerebras

@thsottiaux Small feature request: it would be really helpful if chat input could autosave while you’re typing. Codex crashed on me for the first time, and I ended up losing what I had written. PLEASE SANCTUS TIBO

Build your dreams

ChatGPT Work presents

@Teknium You guys come up with the most random but cool shit. Apple of open source software. Hope you're doing well also btw. Thanks for everything you do. You're a pillar

You can now access and activate your banked resets in Hermes Agent directly with /usage reset when using a codex/openai subscription

@JoePro @suno damn this song is fire wth

You ask for bangers this AM? You will not find a better song about AI. New AI video generated workflow. Next one will be exponentially better. @suno

@JoePro Great post, thanks for not making us have to comment/link/like/subscribe/DM to get the actually skill! BIG! how do you think this would compare to fable/sol design? my apprehension for using these sorts of skills personally is that i feel like it would hinder or limit them

coding 8 projects at once with sol ultra on fast mode. this is how you burn through cash!

@tonbistudio @witcheer Why obsidian vault over gbrain or similar? Btw I’m a member on your YT channel. Great content

model: GPT-5.5 as the main driver, mimo-2.5 pro if I need something fast cheap and good for external projects. memory: built-in and Obsidian vault, experimenting with other plugins like honcho and sybil on individual agents (I've got like 5 over different devices/VPS). interface: Telegram for brainstorming and mobile use, TUI for serious coding work, desktop for when I'm working on a bunch of stuff at the same time (projects is a life-saver) orchestration: kanban for workflows, I use a ton of seperate profiles, basically for each task. Stacking cron jobs with noAgent or wakeAgent gates for automation as well Use a lot of frontend skills and creative skills, as well as MCP servers and webhooks to connect externally. Been actively using /learn to create great skills as well for my workflows

I'm interested in how you're all running Hermes day to day. drop your setup below, I'm mapping what the community reaches for. I'm mostly curious about: - model: your daily driver, plus MoA or a local model if you run one - memory: built-in, an Obsidian vault, or another layer - interface: TUI, Desktop, or a Messaging gateway - orchestration: kanban, delegate_task, subagents in tmux, /goal - the skills or MCP servers you'd miss if they were gone no setup is too small. I'll gather the common patterns and share them back.

@sama hilarious post lmao. gpt 5.6 is a serious work horse. i just cant help being a hype machine for this model. king sam!

still sorta breaks my brain to see our models be good at design finally

@thsottiaux TIBO!!!!!!! King Tibo

Tomorrow might be 8M active user celebration day. Just saying

@thsottiaux I seriously cannot believe how much of an amazing experience this release was. The number of resets, the improvements to the model, the temp removal of 5hr. I’m getting HOT FLASHES! Latest model included with our plans Tons of resets 10/10 Tibo for world emperor

Thank you to the 7M active users who are now using Codex and ChatGPT Work. We have added a banked reset to everyone's account to celebrate the milestone. You can apply the reset in the desktop app or on web and it will replenish the weekly usage for you. Have fun out there.

@logicleaplabs @prasenx @grok Awesome, thanks so much for your reply!

@taymour @prasenx @grok Keep Blender open and treat the viewport as the subscribed surface: the MCP call mutates the scene, the add-on emits state, and the capture shows the delta. The missing piece isn’t another prompt; it’s a visible event loop. Persistent sessions make that legible.

just asked gpt-5.6 sol in cursor to set up blender mcp and make me a realistic floating macbook, then render the whole thing. never opened blender once in my life before today.

@Teknium Sorry to hear this sad news. Get well soon

Hey everyone. I haven't been very responsive on here the last week. My dog, Link, who I've raised since he was a puppy over the last 13 years, passed away yesterday after being in the vet ER's ICU since last Wednesday for heart failure. I put together some of my favorite pics of him to share so you all can see the most awesome animal friend I could ask for. I'll be a bit slow probably through this week too, hope you all can understand 🙏

@thsottiaux KING!

Rest assured that GPT 5.6 Sol will stay in the ChatGPT subscription you pay for. Including Go, Plus and Pro subscriptions. At least until we ship an even better model.

@thsottiaux @AArdvarkErick IT HAPPENED HOLY SHIT

@thsottiaux @AArdvarkErick You can't keep everybody happy! Thank you for everything. SPEAKING OF WHICH.. Please fix 5 hour limits, I want to melt into the ground :D

At different times you complained about speed, code slop, frontend quality, ... With each release we improve and GPT 5.6 Sol is ✅Fast and token efficient ✅Hardcore at back-end dev ✅Great at front-end ✅Does not use useEffect everywhere What is next?

@Tv2Filip @pauloportella_ @thsottiaux damn, do share!

@pauloportella_ @thsottiaux I am currently sitting at 9 resets so I guess unlimited?

Added a banked reset to 500k users of ChatGPT Work and Codex. What’s been happening: - Just released the ability to use a banked reset from web and mobile. Before today you could only do it in the desktop app - Had an issue where less than 10% of users who used a banked reset and it didn’t actually reset. This was during a 2 hour window. - It was hard to find all the exact users where it didn’t apply, so instead granted a banked reset to everyone who pressed the reset button in that 2 hour window. This also gives us the opportunity to validate the new infra ahead of tomorrow Tomorrow we will celebrate our 7M active users milestone and grant the first banked reset across all of our ChatGPT Work and Codex users. We will also release a bunch of updates to the desktop app, addressing a ton of your feedback. Talk soon

@thsottiaux Genuinely want to reach out to the vatican to have you considered as the first living saint

GPT 5.6 Sol being a workhorse is no exaggeration. I was able to complete a 7 hour /goal this evening 🥹now I just ran out of weekly. Santus @thsottiaux 🙏🏻 reset?? 🍗

Every time I see something cool with LLM and someone is showcasing a video, I immediately assume it was done with Hermes. @Teknium @NousResearch continue to innovate!

@filicroval @NVIDIAAI So this is basically you asking it to play the game via pc control right? And then you’re seeing how it performs? If so, very cool

I tested @NVIDIAAI's Nemotron Puzzle 75B locally on my GX10 with 3 HTML game challenges: - Tower Defense 🗼(~ 1m30 at 60 tok/s) - Wrecking ball 🏐(~ 40 s at 60 tok/s) - 2D Roguelite 👾(~ 1m at 60 tok/s) The results are satisfying considering that there was no harness involved (ignoring the shaggy physics of the wrecking ball one)