FiMo ➕

[FiMo ➕ @FiMoTW]

@luongnv89 @skcd42 same, no changes

[FiMo ➕ @FiMoTW]

唯一的缺點是usage消耗太快, 才一個上午就已經消耗9%, 且還要等到六月底才會reset, 這段空窗期實在太久了 💀 @xai

[FiMo ➕ @FiMoTW]

Composer 2.5 快到我根本沒辦法分心, Lock-in!

[xAI @xai]

Composer 2.5 is now available inside Grok Build.

Composer 2.5 is a fast, highly intelligent model that excels on long-running tasks and following complex instructions.

[FiMo ➕ @FiMoTW]

Nemotron 3 Ultra!

[FiMo ➕ @FiMoTW]

Nvidia GTC event 2026

[Elon Musk @elonmusk]

SpaceX Dragon spaceship docking with the International @Space_Station

[FiMo ➕ @FiMoTW]

GG

[FiMo ➕ @FiMoTW]

My Kimi Tokens: Gone.
My confidence: Gone.💀

[FiMo ➕ @FiMoTW]

@_TobiasLee 已經收到, 非常感謝

[FiMo ➕ @FiMoTW]

Just as my Kimi coding plan hit zero, Xiaomi dropped free MiMo tokens. -> It’s time to try MiMo-V2.5-Pro.

Back to work!

Thank you @XiaomiMiMo

[Fuli Luo @_LuoFuli]

Just dropped two open-source models: MiMo-V2.5-Pro (Code Agent, 1T total) and MiMo-V2.5 (Multimodal Agent, 310B total).
Oh and one more thing — we're giving devs & creators 100T tokens on us. Go build something cool 🛠️
🎁 100T Free Token Grant for Builders https://t.co/Sfykf7As2J

[FiMo ➕ @FiMoTW]

@Kimi_Moonshot Congrats! I also feel like Kimi k2.6 is doing a solid job with daily coding tasks.

My only complaint is the speed, I'm on the Kimi Coding Plan, but the output is only around 20 tokens per second. It’s just too slow.

[FiMo ➕ @FiMoTW]

@YakuzaDaddy @MiniMax_AI @Kimi_Moonshot @Alibaba_Qwen @OpenAI @claudeai Agree, minimax m2.7 can only do docs job not for coding

[FiMo ➕ @FiMoTW]

Standard Kimi Code was fine, but a custom setup is a whole different vibe.🔥

@Kimi_Moonshot

[FiMo ➕ @FiMoTW]

Give my @Kimi_Moonshot CLI a new look 😎

[Aryan @justbyte_]

Me prompting claude at 3AM:

[Claude @claudeai]

Introducing Claude Opus 4.7, our most capable Opus model yet.

It handles long-running tasks with more rigor, follows instructions more precisely, and verifies its own outputs before reporting back.

You can hand off your hardest work with less supervision.

[FiMo ➕ @FiMoTW]

Just realized I went a whole day without touching Sonnet or Opus! Kimi K2.6 is actually killing it. My token usage spiked by 40% in just 24 hours. 🤯

[FiMo ➕ @FiMoTW]

試用了 @Kimi_Moonshot K2.6-code-preview 兩天, 雖然沒有數據比較, 但是感覺至少有Sonnet 4.5/4.6 的程度了, 差別在解決問題的速度比較慢一些而已, 但是穩定, 穩定, 穩定

[FiMo ➕ @FiMoTW]

試用了 @Kimi_Moonshot K2.6-code-preview 兩天, 雖然沒有數據比較, 但是感覺至少有Sonnet 4.5/4.6 的程度了, 差別在解決問題的速度比較慢一些而已, 但是穩定, 穩定, 穩定

[FiMo ➕ @FiMoTW]

更新 Trident v1.0.5 了 🔱

- /tri auto — 一口氣從設計做到實作，不用中間手動切換
- /tri models — 便宜的模型寫 code，貴的來挑毛病，避免相同模型陷入盲點
- 視覺維度 — 新增 Visual Quality + Creative Impact，做前端/動畫表現更好
- Skill stacking — 自動載入你的專業 skill 給三個 agent共用

終於不用焦慮 @MiniMax_AI or @Kimi_Moonshot token plan 沒有用完的困擾了哈哈 😂

以下測試影片由 Kimi k2.5 寫code, Opus 4.6審查

[FiMo ➕ @FiMoTW]

Released Trident 🔱
> 開源的對抗式設計審查 agent skill。

你一定遇過：AI agent 寫完程式，自己審查自己，說「沒問題」。結果上線就炸了

Trident 讓三個 AI 角色互相爭論，直到你的工作真正到位。

免費開源，回饋社群，給顆 ⭐，一起讓它變更好

有任何想法都歡迎跟我說！

https://t.co/KZslwps11r

#ClaudeCode #OpenCode #OpenSource #AgentSkills

[FiMo ➕ @FiMoTW]

Had @MiniMax_AI M2.7 build a weather animation, then applied my skill — the result is honestly impressive!

Left is raw output, right is after skill optimization.

> Same model. Better workflow = more potential unlocked 🔥

Same kid, different results when someone's watching 😂

[FiMo ➕ @FiMoTW]

Released Trident 🔱
— an open-source agent skill for adversarial design review.

We've all been there: AI agent writes the code, reviews its own work, says "looks good." Then production breaks

Trident fixes this — three AI roles argue until your work is actually ready.
> One builds, one attacks, one keeps them honest.

Free & open source. Give it a try, drop a ⭐, and let me know what you think — let's make it better together!

https://t.co/KZslwps11r

#ClaudeCode #OpenCode #OpenSource #AgentSkills #AI