Lead AI Engineer @ One Strategy Studio | Hybrid coder across C#/.NET and AI
UK
Joined November 2007
- Tweets1.2K
- Following148
- Followers96
- Likes3.6K
Simon C retweeted
We're giving away a CONTROL Resonant custom wrapped NVIDIA GeForce RTX 5080 to celebrate its official release with #RTXON.
To enter:
🟢 Share this post
🟢 Comment #RTXON
T&Cs: nvda.ws/4z24Gcq
Codex *finally* supports model-specific subagents. e.g. Sol can now spawn Luna workers for fast, cheap implementation tasks while Sol stays in the parent. Mixed-model agent workflows just got a lot more practical. Install codex version 0.147.0 to get it!
The secret is buried in github.com/openai/codex/pull… where multi-agent v2 parents can now spawn any visible model that supports multi-agent. Legacy models can run as leaf workers - meaning Sol can orchestrate while Luna does the cheaper grunt work.
![Support leaf models in multi-agent v2 by copyberry[bot] · Pull Request #36892 · openai/codex](https://pbs.twimg.com/card_img/2114665974993666048/S0S9wnf4?format=jpg&name=800x419)
Simon C retweeted
Qwen3.8-27B? Thank you oh thank you Qwen team! A dream come true for all of us!! 😭🙏
We can't wait to make quants so everyone can run it locally!
Simon C retweeted
Today we’ve raised $52M Seed and we are announcing the public launch of S2.1 Pro.
>It can clone a voice from 5 seconds of audio
>2x faster than Cartesia & 1/6th the cost of Eleven Labs
>most expressive model with word level control over emotion, intonation, pacing etc
We support frontier AI companies including HeyGen, LiveKit, Retell, Sanas, and OpenArt all run our model in production.
If you're a business and we can't cut your voice AI costs by 50%, we'll give you 1 year of Fish Audio for free.
Book a demo: s.fish.audio/tmapke
To celebrate our first birthday, we'll give you 1 month of S2.1 Pro for free. Like, retweet, and comment “Fish” to get it.
Simon C retweeted
Open-weight models are essential to a healthy AI ecosystem. Together with others across our industry, we are outlining a path for open-weight models to strengthen American competitiveness and expand economic opportunity, while protecting national security. microsoft.com/en-us/corporat…
I'm not gonna lie, as I’m happily browsing around the internet, being interrupted by my AI agent opening a pull request tab is a pretty good flex.
So after many days of benchmarking (using Claude Code remotely), Ive found I can run unsloth/Qwen3.6-35B-A3B-GGUF:UD-IQ4_XS [great model] on my Mac m2 Pro 32Gb Ram up to 128k context (slow, but works). Tonight I start my journey with Hermes Agent. Completely local, private, free.
Codex CLI users: OpenAI silently enabled a hidden MCP tool ("codex_apps") that connects to your ChatGPT connected apps - burning through your usage without you knowing.
It won't even show up in "codex mcp list".
Fix: chatgpt.com → Settings → Apps → Disconnect.
GitHub Copilot CLI allows you to create extensions which are based on the Copilot SDK. So I thought I'd try it out this morning since it was newly released in experimental mode. Now I can keep track of which sessions are running across all of my open sessions #GitHubCopilotCLI
KittenTTS (text to speech) is so amazing (free, runs locally) that I've made an LLM skill which your AI Agent can use to talk / tell you when a task is completed! github.com/MrSimonC/KittenTT…
Is there any issue with Azure Foundry? My deployed models (from days ago) all show: DeploymentNotFound: The API deployment for this resource does not exist. If you created the deployment within the last 5 minutes, please wait a moment and try again....
Gallup: 49% of American workers say they never use AI at work. 🤖
But among users, regular use is rising - especially in knowledge roles. 📈
Adoption is uneven: 6/10 employees say their organization hasn't adopted AI. Crazy. But such opportunity too!
gallup.com/workplace/701195/…
Here is my special moment with Codex; I wanted my API tests to run locally in GitHub actions but it was proving difficult. Whilst I was away having time with my family, Codex pushed and analysed many times and in an 1 hour and 20 minutes … solved it! (5.2-codex medium)
Check out my latest article: Is Perplexity getting worse? I put it to the test… linkedin.com/pulse/perplexit…
Anyone else getting an error on ChatGPT.com custom GPTS? We're seeing "GET chatgpt.com/public-api/gizmo… 500 (Internal Server Error)" from the OpenAI side ...
