✝️ Jesus is Enough | I help creatives and believers leverage technology to amplify their art and faith, while exploring the world.
Joined April 2019
- Tweets9.7K
- Following1.3K
- Followers2.9K
- Likes30.9K
🔗 Try it free: davidsandyofficial.github.io…
What should I build next?
Fan-made · Unofficial · Not affiliated with SNEK · NFA
Borrowing an audience works on one condition: the quote has to add something. A one-line take people agree with beats a paragraph that just repeats the post. What's the best quote tweet you've seen lately?
Half the token price and some people already prefer the output. That's the question for most of us: not "is it the smartest," but "is it good enough for the job at a better price." Have you tried Terra yet?
Splitting work across models instead of throwing everything at the biggest one is the AI version of delegating well. Right task, right person. Do you stick to one model or mix them?
The Head of Codex at OpenAI just shared an official tip as GPT-6.1 Terra drops today
The setup: split your workload across GPT-6.1 Sol, GPT-6.1 Terra, and GPT-6 Astra, and stop burning Sol context on tasks Terra can swarm at 340 tok/s
Zero context pollution between sessions: independent subagents execute via plain-text IPC in separate worktrees without burning a single token on conversation history
You just launch Codex like this:
codex --model gpt-6.1-sol --agent adversary --subagents gpt-6.1-terra
here's the split:
→ GPT-6 Astra architects invariant security: audits ~/.codex/agents/adversary.toml, enforces sandbox policies, and signs off on multi-worktree diffs
→ GPT-6.1 Sol executes code fixes: scopes architecture, patches syntax, and merges pull requests whenever a worker reports a completed diff ($2.00 / $10.00 per 1M tokens)
→ GPT-6.1 Terra swarms parallel subagents: delivers near-frontier reasoning on tool use at $1.00 / $5.00 per 1M tokens (50% cheaper than 5.6 Terra, 3x faster than Sol) and burns through AST discovery and test suites at 340 tok/s
→ you stop burning $10/1M Sol context on reading 30 files, cutting agentic swarm costs by 80% while keeping your lead session at a pristine 14k tokens
Astra only steps in at three moments:
→ setting up subagent topology and auth invariant rules before launch
→ evaluating contract diffs when a test breaks twice or reports an API schema update
→ signing off on synchronized multi-worktree PRs before merging to main
Astra architects · Terra swarms · Sol executes
JEV engineering is the same move one layer down: the forks that need no thinker (which worker, which tool, forward or drop) go to JEV in milliseconds, and the big models only see the ones that split
the full tree
> GPT-6 Astra on call as the adversary auditing subagent topology
> GPT-6.1 Terra swarms parallel subagents at $1.00 / $5.00 per 1M tokens with near-frontier accuracy at 340 tok/s
> GPT-6.1 Sol patches code and synthesizes PRs when a subagent signals completion
> Subagents run in isolated git worktrees with zero conversation history dump
> Messages travel as isolated plain text over local IPC with 0 context dump
> JEV routes micro-forks in milliseconds
paste the tree and this prompt into Codex CLI ↓
"Rebuild my Codex setup around this tree:
1. Configure subagent discovery using ~/.codex/agents and set default_subagent_model to gpt-6.1-terra
2. Delegate all worker execution, AST discovery, and test suites to GPT-6.1 Terra ($1.00 / $5.00 per 1M tokens) at 340 tok/s
3. Spawn GPT-6.1 Sol to execute code edits and synthesize PR merges whenever a subagent reports an update
4. Restrict all subagent traffic to lightweight isolated git worktrees with zero history transfer
5. Reserve GPT-6 Astra for invariant authorization and final sign-off before committing
Show me every change as a diff first. No edits until I say go"
↳ developers.openai.com/codex/…
bookmark this, your subagents now swarm without burning lead context
"I am a product guy" is the lesson. A good product gives your marketing something to work with. What came first for you, the product or the audience?
Simple and true. You don't need a quiet room or a perfect morning. A few minutes in Scripture on a busy day still counts. When do you usually fit it in?
"Build the creative app you wish existed, just by describing it." That's the shift: creators going from users of tools to builders of tools. What app would you describe first?
Love this question. Small, daily habits beat big promises we never keep. For a lot of us it's just opening the Bible before opening the phone. What's yours?
Agree. Main Street doesn't need an AI lecture. They need someone to fix one annoying task: the phone, the booking, the follow up emails. Which local business near you needs this most?
Good filter for online business too. The people flashing the lifestyle are usually selling the lifestyle. Look for the ones showing the work, the numbers and the boring parts. Who's someone you trust online and why?
"~9x lower cost per task" is the line small businesses should care about. When AI work gets cheaper, the tasks that weren't worth automating last month suddenly are. What would you automate if cost wasn't the issue?
BREAKING 🔥: OpenAI officially announced GPT-6.1 Terra!
> "At max effort, Terra is over three times as token-efficient as Claude Opus 5.5, and one-fourth its price per token, resulting in a ~9x lower cost per task."
> "Terra is also much quicker, completing tasks over three times faster than both GPT-6 Astra and GPT-6.1 Sol."
I would call it a Berlin DevDay release 👀
Side by side on the same form is the right way to judge a new model. Skip the benchmarks and test it on the boring task you actually do every week. What form would you make it fill out?
i asked GPT-6.1 Terra to show me its computer use alongside Sol.
here’s my prompt:
“I need you to find a long but generic form on the internet and record my screen while GPT-6.1 Terra fills it out in Fast mode. Then have GPT-6.1 Sol fill out the same form in Fast mode and show both recordings side by side."
in the original run of the same form:
Terra took 53.5 seconds vs 3:21.7 for Sol.
that’s a 2:28.2 difference!
that run used accessibility controls, which computer use normally defaults to)
the recorded video uses mouse controls so you can see the cursor.
Most investors disappear after the check clears. Being the "old friends" a founder still calls 10 years later is a brand no ad budget can buy. Same rule works for any small business with customers.
This is how to use AI for research: don't ask for an answer, ask it to pull fresh data and extend the existing work. Then check it. What's a "known fact" in your field you'd want re-tested?
Good reminder that AI models get attacked like any other software, and get patched like any other software. If you build on these tools, keep an eye on the security notes, not just the launch posts.
"Faster at using a computer than what I can do unassisted," from OpenAI's president. If that holds up, the skill that matters is knowing what to delegate. What would you hand off first?
Most people waiting to feel "extraordinary" before they start will wait forever. The work is the part you control today. What's one thing you're putting extra reps into this week?
6.1 Terra being pitched on computer use is the real headline. Chatbots answer questions. Computer use means AI clicking through the boring stuff in your actual apps. Which task would you hand it first?
