CommandCode.ai
@commandcode.ai
2 months ago Grok 4.6 is now live in Command Code GOAT.
💬 0
♻️ 0
❤️ 0
Grok 4.6 is now live in Command Code GOAT. tyty!
first small step along the way to making Atuin sync SO MUCH faster
Our engineering and design team has been testing 26+ side-by-side comparisons across frontier and open models. All runs are public and open source. Benchmark for this demo here: github.com/CommandCode... Tested New DSV4 Pro 0813, Kimi K3, and GLM 5.2 with our FlappyBench
3 models, same prompt with the /design command.
Reviewed gameplay features, UX/UI, and cost.
🔹 DSV4 Pro 0813 → 8/10 · $0.0005
🔹 Kimi K3 → 9.5/10 · $0.0740
🔹 GLM 5.2 → 9/10 · $0.0480 Do you want early access? We're inviting more users over the coming weeks. Sign up to get in: zed.dev/deltadb In Delta, the conversation is a document, and your cursor works everywhere in it. Move through the thread with the same keyboard motions you use on code. When you want to respond to something, put your cursor on it and start typing, nothing in the thread is off limits.
Share a thread with a click and teammates join as first-class participants. They can explore the history, co-edit prompts, or pick up the work days later without wondering whether the latest code was committed or pushed.
In Delta, review happens where the work happened. The agent that wrote the code is still in the thread so when something looks wrong, you ask it instead of reconstructing intent from a diff.
Delta is built on DeltaDB, which keeps your code and conversation in sync for everyone in the thread, in real-time. It works with the git repo you already have. Every edit and conversation is captured between your commits.
Today we’re introducing Delta. Delta is a multiplayer environment for coding with agents and reviewing what they build, powered by DeltaDB.
The first invites to our private beta are going out now.
zed.dev/blog/introducing-delta Our team took a look, and it seems specific to TS 7. Try adding overrides: { typescript: 5.9.3 } to pnpm-workspace.yaml, plus "typescript": "5.9.3" in your root package.json devDeps so Nx resolves it explicitly. Then clear the cache and deploy. OpenRouter is live in Netlify AI Gateway and Agent Runners! Chris Watts (OpenRouter) and Jack Herrington will build with it live on Tuesday, August 18 at 12pm ET, swapping models mid-app without touching the code. Register: luma.com/c37azk4s You’ll hear from builders working on AI agents and the tools behind them, sharing practical lessons from production.
We’re taking over SPIN San Francisco for talks, networking & ping pong 🏓
🎟️ Super Early Bird tickets are $150 until August 24:
pris.ly/builders-da... We’re bringing AI Builder’s Day conference to San Francisco 🎉
On October 26, engineers, founders & AI builders are coming together to talk about building & shipping AI in production.
Talks, demos, real-world lessons & time to meet other builders (1/2) 🧵 20 days left to climb. Join at ntl-back-to-school.netlify.app and see the wall at ntl-back-to-school.netlify.app/showcase. 🎒 The 63 apps split into 35 interactive builds, 11 static explainers, 10 games, and 7 simulators. The rule from day one still holds: whatever makes the idea click counts.
raimanncostigan holds ten apps and a nine-day streak with Page Fault Lab, and Fixiki sits at five after Stock Market Simulator. Day 11 is still ahead for them.
AshAmplifies is the quiet climber. Mental Math landed on Day 10 and kept a perfect ten-day streak alive, unbroken since Day 1, and the tiebreaker lifts him to third. anishsharma is back with RTG after missing Day 10, so his ten apps now ride a one-day streak.
davidvkimball stays level at eleven and hangs an eleventh plate at yahew.com. The Guestbook rebuilds a 1995 Perl CGI guestbook, comments you owned, spliced into the page at an HTML comment. Sign it and it's real. waela is the only builder at a perfect eleven-for-eleven. Today's hand-in is Real Animal or Made Up?, a game daring you to tell nature's weirdest creatures from ones we invented.
🎒 Netlify Back to School, Day 11 of 31. Eleven days in, ten builders have deployed 63 teaching apps, and the race has turned into a question of who hasn't missed a day. The assignment hasn't changed: build an app that teaches one idea, ship it, come back tomorrow. 🧵
What if you could focus on the feature instead of the AI plumbing? KP explains how AI Gateway takes care of model configuration and provider management so you can spend more time building.
In our upcoming webinar, we’ll show how Elicit Research Agent can go from research to a decision.
Join us for the webinar:
Using Elicit to Make More Rigorous Decisions
Thursday, August 20th
9AM PT
Presenter: Kadir Annamalai, Product, Elicit
Save your spot today crowdcast.io/c/ux323o9p85k1