OpenAI might be shipping a persistent agent called o.

2026-09-29 12:58

OpenAI might be shipping a persistent agent called o. Yep, another personal agent. One-letter name, never clocks out. You hand it a goal instead of a message. It runs in a cloud sandbox for hours or days, breaks the task down, fixes its own mistakes, then comes back to you when it's done. Close the session and it's still running. It gets its own email identity, so it can take on async work like your inbox. It's a ChatGPT Pro perk, $100/month and up. $20 Plus users probably don't get it. No word on whether it trickles down. Goes straight at xAI's Grok Bot, but with no standalone app. It just lives inside ChatGPT.

OpenAI released GPT-6.1 Sol.

It scores 75.2% on DeepSWE v1.1, ahead of flagship GPT-6 Astra at 74.8%, for a fifth the price: $2/$10 per million tokens, $0.10 cached input. One task costs $1.50 against Astra's $7.70. Computer use on OSWorld 2.0: 71.4%, 2.1 points behind Astra, roughly a seventh of the cost. GPT-6 Sol held the flagship spot for 7 days. https://openai.com/index/introducing-gpt-6-1-sol

2026-09-30

DeepSeek-harness got a desktop app

. Did not see that coming.macOS and Windows both. Beyond the official key, it now takes a DeepSeek account login, so they're going mainstream. My bet: account subscriptions ship soon. API key access stays open either way, you can still point it at third parties.

2026-09-29

Claude appeal rejected.

Won't refund the subscription either. I'm half convinced the appeal email is itself phishing, because I'm pretty sure they log the IP. DHH came to Shenzhen recently and showed how he uses Claude Code over there. That's the safest way now, honestly: run Claude Code on an overseas machine, connect through Tailscale, never use it locally. So I need to rent an overseas box with a residential IP, and register and log into Claude there.

2026-11-14

Sharing my current mobile only workflow.

Until late September it was the usual setup: parked in front of my laptop, terminal open, bossing pi around, fanning out subagents, then eyeballing diffs myself and filing the PR. Glued to the chair all day, every day. Late September I moved the whole thing to my phone, Hermes runs the show, pi takes the orders. My back is very grateful. Three weeks in, still no regrets. Stack: Hermes on flagship brains (gpt-6-astra / opus-5.5), pi on mid-tier (ds-v4.1-flash / sonnet / terra) with as few plugins as possible, the two talking over a2a. Hermes reads the issues, slices them into small tasks, spins up a worktree for each one, and dispatches a pi into it. When a pi finishes it pings Hermes, who decides: ship the PR or bounce it back for rework. I get back from a full day of being a tourist, or just wake up, and a dozen PRs are sitting there done. Pretty sweet. The catch is how black-box the whole thing is. My emotional arc has been a rerun of early vibe coding, except back then I at least read the diffs. Now even that goes to a flagship model. Boxes all the way down. If you don't have data anxiety, grok-bot and chatgpt-dot basically ship this out of the box, grok-cli and codex built in, native agent-to-agent comms. Downside: your code lives on their servers, everything's moderated, and one ban takes it all with it.

2026-11-14