Yuting Lin

AI Specialist CE@ Google Cloud ☁️ ex IBM 生成AIについて発信中🇯🇵 Opinions are my own 🙌

[Yuting Lin @YuuLin123]

Make Gemini the workhorse and save more:
Claude as the conductor, Gemini (Antigravity CLI) as the executor — shipped as a Claude Code plugin.

Measured: −27%→−64% frontier-token cost at equal eval quality (3/3).

https://t.co/jaJd8HXlYJ

[ClaudeDevs @ClaudeDevs]

A few patterns we frequently use with Fable 5:

Use Fable 5 as an "advisor."

An executor (Sonnet 5) calls Fable 5 for guidance.

Most tokens are billed at the lower executor rate.

[Yuting Lin @YuuLin123]

@Majin_AppSheet 予約購入で昨日届きましたが、Gemniで賢くなってて良い感じです！

[Yuting Lin @YuuLin123]

構成・実測・使いこなしは記事に詳しく書きました📝
📝 Blog: https://t.co/pfuInc3sHF
⭐ MIT のコミュニティプロジェクト:
https://t.co/RXcuj45EsC

[Yuting Lin @YuuLin123]

Claude vs Gemini という論争をよく目にするが、両方を協働させたらより強力になる。

Claudeが指揮（設計・検証・レビュー）、Geminiが力仕事（実装・テスト・検索）——Claude CodeからAntigravity CLIを動かすPluginを作った。

大規模ビルドで Claude Opusコストを同品質のまま 27〜64% 削減。

[Yuting Lin @YuuLin123]

価値はコストだけじゃない：Vertex AI Search で社内データ grounding、Web/deep research、クロスモデル検証。
Gemini の検索や Agy CLI のハーネスを Claude Code から使えるようになる。

[Yuting Lin @YuuLin123]

It's MIT and early-stage — contributions very welcome 🙌
Especially keen on: more A/B runs (n>1), new SDLC recipes, support for other CLIs.
Open an issue, send a PR, or just ⭐ if it's useful — feedback shapes where this goes.

[Yuting Lin @YuuLin123]

Everyone argues Claude vs Gemini.
I made them work together.
Claude Code conducts (design, verify, review). Gemini does the grunt work (implement, test, search) — from one cockpit.
On a large build it cut my frontier (Opus) cost 27–64% at equal quality. 🧵

[Yuting Lin @YuuLin123]

Try it (needs the Antigravity CLI + Claude Code):

▎ /plugin marketplace add yuting0624/antigravity-for-claude-code
▎ /plugin install antigravity@antigravity-for-claude-code
Repo (MIT) + measured A/B + demo 👇
https://t.co/jaJd8HXlYJ

[Yuting Lin @YuuLin123]

Anthropicが米政府の指示でFable 5 / Mythos 5を緊急停止。
理由は"狭い範囲のジェイルブレイク 1件"らしいが、Anthropic側は「誤解」として復旧を目指すと表明。
GCPなど3P経由も含め影響が出る見込み。

frontierモデルの輸出規制がここまで来たか…という時代の節目を感じる。

[Anthropic @AnthropicAI]

The US government, citing national security authorities, has issued an export control directive to suspend all access to Fable 5 and Mythos 5 by any foreign national, whether inside or outside the United States, including foreign national Anthropic employees.

The net effect of this order is that we must abruptly disable Fable 5 and Mythos 5 for all our customers to ensure compliance.

Access to all other Claude models is not affected.

We apologize for this disruption to our customers. We believe this is a misunderstanding and are working to restore access as soon as possible.

Read our full statement: https://t.co/bwn0sximKZ

[Yuting Lin @YuuLin123]

Claude Fable 5、Gemini Enterprise Agent Platform でも使えます🎉

ただし Model Garden の有効化だけだと 403 エラーになります。Mythos クラスのモデルは Anthropic へのデータ共有オプトインが必須で、コンソールに UI がなく API 経由でのみ設定可能。

手順を記事に追記しました👇
https://t.co/lZivJAG9Nz

[Claude @claudeai]

Introducing Claude Fable 5: a Mythos-class model that we’ve made safe for general use.

Its capabilities exceed those of any model we’ve ever made generally available.

[Yuting Lin @YuuLin123]

@NVIDIAGeForce I want a nice GPU to test local LLMs!
#RTXPowersPlay

[Yuting Lin @YuuLin123]

こういうの待ってました😭
もっと褒めてあげてください

[たぬ @AIエンジニアWWWWWWWWWWWW @t2aman1e]

俺のGemini優秀すぎてヤバい