Indra Gunawan

@bunjavascript @cirospaciari In the next version of Bun: Native Clickhouse please

In the next version of Bun node:sqlite is fully implemented and passes 100% of Node.js’ test suite

@nahid_pro09 NOT FOUND: integrate.api.nvidia.com/v1

you can run over 100 high-performance ai models for "FREE" using nvidia's nim infrastructure without paying for hourly gpu rentals or expensive cloud credits😲 most developers are still burning through their runway on massive cloud bills, but nvidia has opened up their production-grade stack to members of their developer program for zero cost takes 5 minutes to set up what you get inside the nim catalog: -100+ curated, frontier-grade models including llama, mistral, gemma, whisper, and deepseek -the exact same enterprise inference stack used for nvidia's internal production -lightning-fast, low-latency, high-throughput endpoints -full openai-compatible api format for drop-in migration what you are replacing: >colab pro and hourly gpu rental fees >per-request charges from premium api providers >the headache of maintaining self-hosted gpu clusters >expensive cloud inference endpoints full setup guide (nvidia nim implementation): step 1: join the developer program ->sign up for free at: build.nvidia.com/explore/discov… ->no credit card or identity verification is required step 2: generate your key ->navigate to your account settings and generate your personal nvapi-key step 3: plug into your client ->set your base url to :[integrate.api.nvidia.com/v1] ->drop your new api key into cursor, zed, or any openai-compatible framework step 4: scale your agents -leverage the generous free-tier rate limits to run your agent swarms, tool-calling loops, and multi-model pipelines -monitor your usage via the dashboard to keep your prototyping within free limits important notes: ->this is a developer-tier access designed for prototyping and research, not massive commercial scale ->you get an initial pool of inference credits on signup, which can be extended via request ->it is the cleanest way to run production-grade inference without the overhead of enterprise billing already spending a fortune on gpu hours? skip straight to step 1 and move your prototyping to nvidia nim to bring your marginal costs to zero this is the single best way to build agentic swarms on professional-grade hardware for free right now bookmark this and claim your access before the terms shift or capacity limits tighten

@rough__sea Agreed! Long live Opus!

i expect almost all software is about to be rewritten from scratch

@FapSpirit @matteocollina Because Watt is using multithreading

@matteocollina Why Watt has 2 CPU, while the rest has only one?

Bun is fast, until latency matters for Next.js. We benchmarked the same Next.js app across Node.js, Deno, Bun, and Watt (our multi-threaded Node-based runtime) under identical load on AWS EKS. Throughput looked fine across the board. Latency told a very different story. 🧵

@Dr_Huey cuss beli, where to buy?

I’d sell my kidney for this walahi

@fypdonk @bravobeacukai wkwkw ribet ya

@bos5ig @bravobeacukai bawa terbang ke LN terus balik ke indo om klo mai daftar imei ke becuk 😂

@bravobeacukai saya beli iPhone dari marketplace, mau mendaftarkan IMEI apakah bisa dibantu? Dokumen lengkap bisa saya siapkan: * invoice * bukti pembayaran * link pembelian * npwp

@bravobeacukai @bravobeacukai halo, sekarang status sudah tidak terdaftar lagi, signal tidak aktif. Bagaimana cara daftar dan bayar taxes?

@bos5ig Halo kak, untuk HP pembelian dalam negeri, tidak perlu lagi didaftarkan IMEI nya ya, karena setiap HP yang masuk secara legal pasti sudah terdaftar IMEInya

@bravobeacukai @bravobeacukai Halo, masalahnya tiap 3 bulan signal mati, complain ke seller baru aktif lagi, sepertinya mereka pakai via turis, saya mau aktifkan permanen, bagaimana caranya?

@bravobeacukai saya beli iPhone dari marketplace, mau mendaftarkan IMEl apakah bisa dibantu? Dokumen lengkap bisa saya siapkan: - invoice - bukti pembayaran - link pembelian - npwp

`deno compile` your program with web workers into a single binary

@jarredsumner Just reload entire page? or HMR (partial patch update?)

This is hot reloading on the server

bun --hot

Bun should be listed here soon @jarredsumner blog.cloudflare.com/introducing-th…

@jacob_d_kahn and I would like to share Shumai - a new ML framework for server-side JavaScript (using bun) It's simple and quick! 😁 github.com/facebookresear… (benchmarks below, WIP) #javascript #MachineLearning

Dear @Oracle, Please release the JavaScript trademark! tinyclouds.org/trademark

OTP: Oalah Takkiro Password

v1.25.1 has been released with many bug fixes related to the experimental npm support & new http server github.com/denoland/deno/…

@jarredsumner unetworkingab.medium.com/bun-can-become…

One of my favorite criticisms of Bun “Bun doesn’t add anything new to the JavaScript ecosystem” Very excited to prove those people wrong

I think Bun needs to start prioritizing some new & missing features

The future of farming. Source @gigadgets_