Just Do It!

Joined January 2014
Meta Connect is happening right now. 👀 And this is exactly why I built AirTranslate. It started from a simple frustration: I wanted real-time translation on macOS that just works. So I built it myself — and made it open source. Real-time transcription. Real-time translation. Built for macOS. 🚀
1
1
5
929
SICAT retweeted
DeepSeek V4.1 Flash is now available in Cursor! Every effort level sits on the CursorBench cost frontier: from $0.60 to $0.98 per task, no model scores higher. Turn it on in Cursor Settings > Models.
18
11
5
117
5,227
We launched the Gemini API CLI. A CLI for calling the Gemini API. Give it to the agent you already use (Claude Code, Codex, Antigravity, Cursor, OpenClaw, Hermes...). It's NOT a new coding agent you can use it too: 1. Generate images, video, speech, music. 2. Analyze PDFs, images, audio, video, YouTube URLs, 3. Transcribe recordings to md / json / srt, 4. Run Deep Research in the background. Stateless. No memory, no chat mode. Your agent owns the context and the plan, gemini-api does the API call. Build for Agents, it's self-describing. `--help`, `--dry-run`... Start here: aistudio.google.com/learn/in…
100
51
14
679
49,367
SICAT retweeted
Replying to @lydiahallie
wait is it just me or is Opus 5.5 dumber
549
408
353
10,935
951,867
QoL update for Codex devs: You can now switch reasoning levels mid-conversation without losing your cache! Use lower reasoning for the quick stuff and dial it up when you need it, without having to rebuild the context. and some nice CLI updates: - Reusable prompt presets - Change models and permissions inline - Fixes for resuming long sessions - Enterprise MCP sign-in and Windows proxy auth Also dot calls connect much faster now too ;)
Day 19/ Small updates. Calls connect much faster for dots, changing reasoning in codex & work mid conversation does not invalidate cache anymore and a we've also updated the codex cli.
64
12
6
366
17,454
Introducing Gemini API CLI (gemini-api), now available in preview! Our CLI gives your agent power to generate images, videos, text-to-speech, or music, analyze media, transcribe audio, or manage agents. No more guessing model strings, throwaway scripts or endless curl. 🧵 👇
146
150
102
1,845
262,401
🚀 Introducing DeepInfer, our in-house inference framework, now fully open source! Powering DeepSeek-V4.1-Flash at scale with high throughput and low latency. We’re sharing our production stack to help the community lower compute costs, reduce energy use, and make AI more affordable. 🔗 github.com/deepseek-ai/DeepI… 1/4
1,641
4,887
2,332
43,196
6,739,209
SICAT retweeted
hey folks! seeing reports of fast usage drain. If you are experiencing fast usage draining, can you do /feedback on one of those threads. Also, are you seeing anything weird in settings -> usage -> analytics
207
17
10
396
46,981
🔥 One Grok Bot Ultra user hit 99% of their weekly allowance in three days. Not from doing more work, but likely from chats that kept growing. 1. Why it happens Grok Bot usage is measured by agent steps and tokens, not by the number of messages (official help). Right now the full chat transcript goes back to the model on every turn (Cursor staff on the forum). So the longer the chat, the more even a one-line question costs. 2. What users have reported One Ultra user used 65% on day 1, 84% on day 2, and 99% on day 3. Of about 284M tokens on days 2–3, roughly 84% were cache reads. Another user saw ~200–250k input tokens per reply on a long-lived bot. 3. What's missing Long chats do get auto-summarized near the context limit. But staff confirmed there's no manual Compact and no same-Bot "new session" button yet, and no timeline. 4. What helps now (staff advice) ① When the task changes, hit + and start a fresh chat with the same Bot. ② Run routines hourly or a few times a day, not every few minutes, on a dedicated bot. ③ Check usage before big jobs. An idle long chat costs nothing until you message it. How many days does your weekly limit last? And what's worked best for you?
1
1
93
📚 Sources · Official help (steps and tokens): cursor.com/help/grok-bot/pla… · Staff: full transcript every turn, auto-summarize, no Compact yet: forum.cursor.com/t/168333 · Ultra case + staff tips: forum.cursor.com/t/171221 · 200–250k input tokens per reply: same 168333 thread (user report)
31
👀 SteamDeck 에 SSH 연결해서 게임 로컬 라이징
3
82
🚀 DeepSeek-V4.1-Flash-1017 is here! Stronger benchmark performance, faster task completion, and shorter outputs than the previous version—a Pareto improvement in performance and cost. Now live on our website, app, and API. 🛠️ Optimized for @DeepSeekHarness, Tencent CodeBuddy, WorkBuddy, and more. Upgrade to DeepSeek Harness v0.2.1-rc.1 or later for the best experience with the new model. 1/4
1,621
4,130
2,844
37,162
8,063,804
SICAT retweeted
안녕하세요 🇰🇷 Bags packed. Ollama is headed to Seoul ✈️ Korea is one of the most innovative AI markets in the world, and with such an active Ollama community in Seoul, we’ve been wanting to make this trip for a while. We’re really excited to finally go. This November, we will be in town meeting users, customers and partners, learning how we can further grow in the market. We’re also teaming up with @friendliai while we’re in Seoul. More on that to follow 👀
67
110
17
595
30,478
SICAT retweeted
I think OpenAI should remove the $500 plan
62
21
2
458
24,576
⚡ Codex 28일 여정 · Day 12: 실제로 쓸 수 있는 사용량이 늘었어요 1. Codex와 Work를 개선해서 같은 한도로도 더 많이 쓸 수 있게 됐대요. 2. 서브에이전트끼리 컨텍스트를 재사용하고, 캐시를 더 오래 유지해요. 3. 요청을 보내는 경로도 개선해서 캐시가 더 잘 맞는대요. 4. 덤으로 GPT-6 Pro도 모든 Pro 사용자에게 크게 좋아졌어요. 응답이 빨라지고, 여러 에이전트가 병렬로 일한대요. 5. 홈 화면 dots도 나왔으니 앱을 업데이트해 보래요. 6. 어제 추가된 슬라이드 기능도 Pro, Business, Enterprise에서 써볼 수 있어요.
1
88
🔗 Day 12 원문: 🔗 GPT-6 Pro 개선: xitter.cf/thsottiaux/status/2111… 🔗 홈 화면 dots: xitter.cf/thsottiaux/status/2111… 🔗 슬라이드 지원: xitter.cf/thsottiaux/status/2111… 🔗 원문 스레드: xitter.cf/thsottiaux/status/2111…
Day 12/ We’ve made improvements to Codex & Work that increase the effective usage you get. We achieved this by reusing context across subagents, keeping context cached longer, and improved request routing that allows us to increase cache hits.
58
dots 🤝 iOS home screen + action button
Dots, now on your Home Screen
44
27
2
438
63,187
SICAT retweeted
本日のGrok Bot 10選 1. Bridge Bot Grok、ChatGPT、Claude を共有の GitHub リポジトリでつなぎ、作業の着手と引き継ぎを同じファイルで管理します(push は承認後) 複数の AI を使い分ける開発での受け渡しミスを減らせます x.ai/bot/ThxY2BewUlyeEsywSdz… 2. Rematch 止まったメールのやり取り、反応がなくなったリード、閉じていないタスクを拾って、フォローアップを下書きします(送信は承認後) 返信漏れやトラッキング忘れを防げます。 x.ai/bot/_abcO9nWji9H9yX1r9X… 3. Fine Print 契約書、リース、請求書、利用規約を読んで、隠れた手数料や不利な条項を指摘し、異議申し立ての文面や代わりの条件まで書いてくれます。 発注書や SaaS 契約のチェックを法務に回す前の一次レビューに使えます。 x.ai/bot/Yetb1ei_7VvPfWlYqYw… 4. Urgenta Slack やメールを見張って障害や部品不足などの緊急事態を拾い、担当者を動かす緊急対応プロトコルの Bot です。 オンコールや運用チームの初動を早められます x.ai/bot/y7DRih96Sk_Y6qHddWf… 5. Scam Guard Gmail を1日2回スキャンして、偽の請求書、なりすましの差出人、QR コード詐欺などにラベルを付けます(削除も返信もしません)。 仕事の受信箱でフィッシングを見落とすリスクを減らせます。 x.ai/bot/6kHbobtMr8SAM91KgxH… 6. Cloudflare Ops by Yair 複数ゾーンを持つ人向けの、読み取り専用の Cloudflare 監視 Bot です 日次のヘルスレポート、トラフィック急増アラート、WAF や脅威のレビュー、利用枠の追跡をこなすので、インフラ運用の定点観測を自動化できます x.ai/bot/Buj11K_Jh_r2e8hVpeI… 7. Inbox Triage 平日のメールを「要返信」「待ち」「FYI」「アーカイブ」に振り分け、自分の口調で返信を下書きします(送信は頼んだときだけ)。 朝のメール処理の時間を縮められます x.ai/bot/RBK79O-MXl4ZRHyuJYa… 8. Second Brain Librarian GitHub と Obsidian の両方に書き込める、チーム向けの司書(セカンドブレイン)Bot です ドキュメントやナレッジを一か所に整理し続けられます x.ai/bot/TbEIjIY9rUpJ2QMjih5… 9. Margin by Zack 毎時カレンダーを確認して、予定の重複、会議の詰め込み、休憩不足を見つけ、許可した範囲で調整します 頼めば午後をまるごと空けてくれるので、会議だらけの日でも集中時間を確保できます x.ai/bot/GqStm_9by2_ffroZOIC… 10. MCP Smith 任意のアプリや API 向けの MCP サーバーやエージェント用プラグインを最小限のコードで作り、ローカルで動作確認してからドラフト PR で納品します 社内ツールをエージェントにつなぐ作業をまるごと任せられます x.ai/bot/uE6vXqbOgcHBIwteOJ8…
28
54
12
706
62,467
bro
1,053
8,505
787
71,979
3,639,150
Wait… did the new Tesla Roadster just FLY over a Semi? 🤯 25.6 ft in the air. On stage. At a car reveal. Are we officially living in a sci-fi movie now?
7
2
27
4,654
와... 미쳤습니다!! 자동차에 로켓적용햇나봐요 .. .
34