mickey

I do Automation, AI engineer, Saham, Crypto.

[mickey @mickeyxdev]

TPIA seru banget gepuk2 an nya hari ini gaes . Ada yang sangkut jg 😭 ??

[mickey @mickeyxdev]

$HYPE mencetak ath lagi hari ini. Makin rame nonton posisi short whale @loraclexyz yang sudah membengkak ke size 115m$ (hampir 2T) dan liquidation price nya di sekitar 90$. Floating loss sekitar 36m$ 🤪
tapi doi juga masi ada posisi spot hype sekitar 57M$.

Apakah Hype bakal seperti bnb run lari ke spot 3 besar di next bull run ?

https://t.co/hS7nh9KSw3

[mickey @mickeyxdev]

Disuru pake opus trus, sonnet sm haiku napa gak pernah diupdate lagi ya ?

[ClaudeDevs @ClaudeDevs]

Opus 4.8 is live in Claude Code today.

A few things worth knowing: 🧵

[mickey @mickeyxdev]

@kompascom @realDonaldTrump bring this guy to white house please

[mickey @mickeyxdev]

@LambeSahamjja Hyperliquid bang

[mickey @mickeyxdev]

Sedikit tips optimisasi cost buat yang vibecodes, kalo kamu pake posgree db (supabase, neon, etc) dan costnya uda lumayan. mungkin uda saatnya coba explore pakai redis buat ngecache data yang bisa di cache. bisa coba di upstash, harganya terjangkau, dan ini ngebantu gw ngurangin cost posgree sampe 80% dan ngurangin workloadnya. minta ai dan setup nya tidak terlalu ribet, asal ngerti mana yang bisa di cache dan kapan buat invalidate cache nya 😅

[mickey @mickeyxdev]

@dojjunn @saptaipb uda loss gede, gk bisa cutloss lg, gimana gak ngamuk

[mickey @mickeyxdev]

@davidsunarna blom lagi server dan embel2 nya bang

[mickey @mickeyxdev]

@gusadib so far cukup bagus bang dia reasoning nya, result nya juga cukup akurat, baru pake bugfix dan modify2 code saya. kalo kimi agak jarang saya make jadi kurang tau jg kalo dibandingkan.

[mickey @mickeyxdev]

Baru subscribe Mimo max plan dan siap mulai 30 days challenge.

Punya 82B Credit (not token).
Targetnya simple: make every penny worth and burn this 82b credit to 0 😆.

will build, eksperimen, ship, repeat.
Let’s see bakal jadi apa di hari ke-30🙆‍♂️

[mickey @mickeyxdev]

@bridgemindai gemini 3.5 flash is opus 4.6 with dopamine. it 3-4x faster.

[mickey @mickeyxdev]

@fajarsentosabuz Pake llm apa bang. Suruh llm ny aj bantu setup lwt terminal

[mickey @mickeyxdev]

Hampir 1 bulan uda pake vps ini dari tencent cloud, so far blom ada problem run hermes + openclaw barengan dengan cost < 1$ perbulan 😄, dan chat dari discord 24/7. Buat LLM nya bisa pakai open router yang free model, tapi lebih reliable pake yang berbayar misal opencode GO (5$ first month, 10$ next/month). so far usage opencode go masih 50%, default pakai deepseek flash V4, kalo minta hermes nya coding fitur bisa pakai deepseek pro V4, ato kimi. Di vps nya bisa juga install claude cli, lalu hermes nya bakalan run claude opus 4.7. Kalau kuota masi banyak, biasa bakal suruh hermes run auto bikin fitur/bugfix, tengah malem sampe quota claude nya habis, besok paginya tinggal review pr nya. Waktu afk juga bisa minta hermes atur todo, docs, pengingat dll, minta dia code dan bugfix dari discord. Overall cukup worth, buat vps tencent ini👍👍.

[Yand @andriyand__]

@mickeyxdev @0xyunss Turor bang 😭🙏🏻

[mickey @mickeyxdev]

@fajarsentosabuz Oh ud sy open dm ny

[mickey @mickeyxdev]

@fajarsentosabuz Bisa bang dm aja

[mickey @mickeyxdev]

@fajarsentosabuz No issue sih so far kalo pemakaian pribadi, chat reply ny cepet jg lwt discord. Tappi Kalau dibuat untuk hosting web sprti vps umumny (host website, service api,dst) mungkin latency nya pengaruh

[mickey @mickeyxdev]

Emang dari dulu Xiaomi ini bikin SEMUA, dari HP, kamera cctv, smarthome, dispenser, tv, vakum cleaner, sampai mobil listrik SU7. Saya pribadi udah pakai beberapa barangnya, kualitas oke dan harganya ramah kantong, puas banget.
Sekarang mereka juga bikin kejutan di LLM juga, dan tentu saja harganya miring dipotong 99% 😆
bakal coba MiMo buat Hermes agent, so far cukup bagus. Nanti gw bakal kasih review lagi setelah pakai beberapa minggu.
kalo kalian juga pakai produk Xiaomi apa aja nih?

[Xiaomi MiMo @XiaomiMiMo]

🚀 Better inference efficiency, lower costs, broader access.

MiMo-V2.5 Series API pricing is now permanently reduced — by up to 99% compared to previous pricing.
✨ Unified pricing across all context lengths.
MiMo Token Plans have also been upgraded:
• 5–8× more usable tokens at the same price
• Simpler and more transparent billing rules
🎁 As a thank-you to current users, all current Token Plan credits will be fully reset.
🎧 MiMo-V2.5-TTS remains free for a limited time.
⏰ Effective May 26 at 6:00 PM PDT.

These improvements are powered by continued inference optimization and serving efficiency upgrades across the MiMo stack.
🛠️ We’ll also publish a detailed technical blog on the inference optimizations later — stay tuned.

[mickey @mickeyxdev]

@indrawxyz @Alibaba_Qwen Pake opencode go pak mantap limit ny dan bisa coba2

[mickey @mickeyxdev]

Sukurlah ada llm model chinese, biar pada nurunin harga jg llm model yang dari US 🙏🙏

[BridgeMind @bridgemindai]

Chinese AI models are dropping prices so fast they're practically giving them away.

And there's a reason for that.

MiMo V2.5 just permanently cut API pricing by up to 99%.

Last week DeepSeek V4 Pro made their 75% discount permanent.

When two major Chinese labs permanently slash pricing in the same month, that's not generosity.

That's desperation.

Nobody was paying full price.

GPT 5.5 and Claude Opus 4.7 charge premium prices because developers actually use them in production.

These models are racing to the bottom because they can't compete on quality.

Cheap and unused is not a business model.

[mickey @mickeyxdev]

@jackfriks @kenn_ronin @marclou I remember you were really active in the Cardano ecosystem back then and I used to watch your videos. Super happy to see how successful you've become man