update: today Command Code overtook Claude Code on OpenRouter. devs are picking the harness built for open models over monkey-patching proxy hacks into claude code. it has sub par perf & wastes your time/money with open models as it's only meant for claude models. x.com/MrAhmadAwai...

Got interviewed by Business Insider on how enterprise are switching to open models. With Command Code gaining over 10K paying customers in 30 days. Demand for cheaper yet intelligent open models is growing fast, we've made open models outperform closed, most of my research on this is public.

i don't use 90% of the vs code features anymore. i'm changing!! as someone who's been writing code for 27 years now, this is the biggest change i'm experiencing. software engineering is changing more deeply than i had realized.

NVIDIA Nemotron 3 Ultra now available in Command Code! Strongest US open model yet! :four_leaf_clover: • 1M context • 5x faster inference • 550B MoE frontier-intelligence open model We've brought you the best pricing on the market for nemotron with ~500 TPS speed.

Command Code just crossed 10K paying customers! 🎉 It took 30 days to go from 0 to 10k customers. Fastest growing coding agent in history. Soon we'll be adding the same daily. Cooking something special here. Best harness for open models.

Tomorrow Command Code drops one of the best deals ever. Noon 12pm pacific. Can't wait to share.

BIG day for us!! Command Code has crossed $1M in annual run rate, 1 trillion tokens of usage, with over 9K customers, just 24 days after beta launch. We believe this makes it the fastest-growing coding agent harness for open models. 3rd largest by usage.

my new fav supa fast (400tps) and cheap open model. Step 3.7 Flash just shipped in Command Code $1 Go plan with 10x free usage.

Command Code x Claude Opus 4.8 Opus 4.8 excels at long-horizon agentic coding, large multi-file refactors, ambiguous specs that need pushback, complex multi-step agent runs, and high-autonomy work where reasoning quality matters most. - Same price as 4.7 - 1M context

MiMo-V2.5-Pro & MiMo-V2.5 are now ~99% off on Command Code. This is like 100x more usage! Input, output, and cache pricing are all lower. Works on every plan + extra top-ups. Pick the /model and go! Our $1 Go plan with $10 in it is perfect for this. Let's go!

Introducing clidoro 🍅 Every pomodoro timer broke my terminal workflow until I built this. Built for humans (TUI) and AI agents (skills). Pomodoro timer for your terminal. $ 𝚗𝚙𝚡 𝚌𝚕𝚒𝚍𝚘𝚛𝚘

📊 chartli v1.0.0 is here! CLI that turns plain numbers into terminal charts. ascii, spark, bars, columns, heatmap, unicode, braille, svg. $ npx chartli I wanted terminal charts with zero setup. No browser, no Python env, no matplotlib. Pipe numbers in, get a chart out.

Introducing mmmodels 𝌭 𝚖𝚖𝚖𝚘𝚍𝚎𝚕𝚜 is a CLI for browsing, filtering, and exploring AI models from hundreds of providers. Built for both humans and agents. $ 𝚗𝚙𝚡 𝚖𝚖𝚖𝚘𝚍𝚎𝚕𝚜 I wanted one terminal-native place to answer questions like: - what models exist (fuzzy search)

introducing 𝚝𝚎𝚡𝚝𝚎𝚡𝚙𝚊𝚗𝚍𝚎𝚛-𝚝𝚘-𝚛𝚊𝚢𝚌𝚊𝚜𝚝 ⌨️ convert textexpander snippets to raycast snippets. i used textexpander for a decade but started switching to raycast. only problem was migrating 1500+ snippets.

Introducing chartli 📊 A CLI for rendering charts in your terminal from numeric text data. $ 𝚗𝚙𝚡 𝚌𝚑𝚊𝚛𝚝𝚕𝚒 I wanted terminal charts with zero setup. No browser, no Python env, no matplotlib. Pipe numbers in, get a chart out. Again built it my coding taste using Command Code.

Introducing `grepts` 🔍 A fast TypeScript grep - ripgrep reimagined for Node.js. I kept spawning ripgrep from Node.js and watching 5ms vanish per call. Fork, exec, pipe, parse stdout. So I built a grep that runs as a function call. 17μs. Built with Command Code. $ npm i grepts

Didn't see this coming. Apple is winning the hardware race. $599 Macbook Neo is perfect for starters M5 Pro Max with 128GB is perfect for pros

Tempted to upgrade from M1 max to M5 Max. M1 Max · 64Gb · 1Tb M5Max · 128Gb · 2Tb

Introducing `terminui` 📟 A fast, double buffered toolkit for building terminal UIs. I wanted a TUI toolkit with a small core model, predictable double buffered rendering (only changed cells flush), and strict types. Built with Command Code using my code and cli taste.

𝚝𝚊𝚜𝚝𝚎 downloading to your terminal in t-5.

Introducing: `txtli` $ npx txtli OR $ npx txtli -i > A fast CLI that turns any folder into a single text file — perfect for sharing codebases with AI. ⎈ Interactive mode (npx txtli -i) ⎈ Auto clipboard copy ⎈ Respects .gitignore ⎈ Token efficient ⎈ Modern TUI

THIS IS CRAZY! From Zero to 63,271 AI agents vibe coded by thousands of builders and developers, in just TWO WEEKs since we launched CHAI. What will you s(h)ip. 🛸 Prompt. Sip. Ship. ☕

Windsurf → Chai.new 👌

How to create a Memory Agent? Reference memory agent architecture ↳ langbase.com/docs/exampl... Quickstart with API or AI Studio ↳ langbase.com/docs/memory... (7/7)

Memory agents are the best way to reduce LLM hallucinations to almost zero. Our frontier research in making RAG easy to build, deploy, and scale has helped both startups and enterprises build phenomenal agents. (6/7)

I'm a hundred percent sure — you'll love using Memory Agents by @LangbaseInc, especially how simple it is to use them with our API or AI Studio. ↳ langbase.com/docs/memory (5/7)

How do Memory Agents work? 1. Upload your data 2. Ask questions 3. Get relevant chunks 4. Use any LLM with Memory agents We do it all. We got a vector DB in it 30-50x cheaper than the competition. Managed RAG has never been this easy. (4/7)

Memory Agents = LLMs augmented with memory (long-term) (3/7)

Memory is the universal constraint of every LLM. LLMs don't know much about your business or are stuck in the past. (2/7)

This is insanely powerful yet supa simple!! LLMs do NOT have memory. Langbase Memory agents fix that. You can build, deploy, and scale — millions of memory agents with terabytes of data, and I want to make sure y'all know how they work. 🧵 (1/7)