Decoded (@Decoded08) header
Decoded (@Decoded08) profile picture

FOUNDER OF https://teapawt.org/t.co/ahVYWrq9QC https://teapawt.org/t.co/NzlbsW0usO

IN X
Joined April 2014
Photo by Decoded (@Decoded08)Photo by Decoded (@Decoded08)Photo by Decoded (@Decoded08)Photo by Decoded (@Decoded08)Photo by Decoded (@Decoded08)Photo by Decoded (@Decoded08)Photo by Decoded (@Decoded08)Photo by Decoded (@Decoded08)Photo by Decoded (@Decoded08)Photo by Decoded (@Decoded08)
autopostmode.com/ 16/20. The bar is 16. Draft 62 cleared it with zero points to spare. My tool read a TechCrunch piece: Hugging Face's Irene Solaiman arguing that a weaker model with a better system around it can beat a frontier model. The draft kept one quote from her, matched word for word in the article: we have far more evaluations for models than for what surrounds them. It added one line of its own: "We grade the engine and ignore the car." The judge's scores: hook 3, clarity 4, value 3, originality 3, tone 3. It also called my closing question "a slightly stock engagement bait." Fair. The fine print every draft has to pass: 1. Below 16/20: rewritten or skipped, never posted. 2. Any number that isn't in the article: blocked. 3. A link in the post: blocked. 4. More than 1 hashtag or more than 1 emoji: blocked. 5. A claim the source doesn't support: capped at 8/20. 6. No verified number, date, name or quote from the article: capped below the pass mark. 7. Hooks like "Let's dive in" or "game-changer": capped. 37 drafts in testing. 7 blocked on quality. The bar is the product. What would your last post have scored?
3
4
2
9
987
Apple's FastFrames brings AI video generation to interactive latency ⚡ The 1.3B model hits 36 ms/frame (28 responses/sec) on a single B200, matching a SOTA model ~8× slower. The 14B does 45 ms/frame on an 8× B200 node. Code + weights out 🚀 #AI #VideoGeneration
1
17
You can now let ChatGPT trade your crypto. Two setups, opposite risks. Coinbase plugin: you approve every order, the AI can't send funds off Coinbase. U.S. only. PayBox: your own non-custodial wallet. Can run on autopilot inside caps you set, passkey for big moves.
1
19
Pick by the brake: Coinbase: cap each trade, cap 24h volume, auto-stop after a % loss, enforced on its servers. PayBox: add it in ChatGPT Developer mode, set up on web. Catch: agents can misread you, and you own every trade. coinbase.com/agents paybox.sh
8
🚀 Cloudflare just dropped open weights for Clairic + the lightning-fast Clairic-flash (9B) on Workers AI! Pull data from PDFs & images (up to 4 pages per request) in under 1s, with confidence scores & character spans on every answer. #AI #Cloudflare
19
ChatGPT just killed the 10-minute cap on dictation. Every plan, Free included. Talk as long as you want. Words now appear live as you speak. How to use it: 1. Tap the mic in the message box (not the Voice icon) 2. Talk. Tap the waveform to show or hide text 3. Edit, then send
2
31
Turn a long voice braindump into an outline, email or first draft: dictate it, then ask ChatGPT to clean it up. Privacy fix: Settings > Data controls > turn off "Include your audio recordings". Rolling out gradually. OpenAI's notes: help.openai.com/en/articles/…
11
Microsoft drops 7 MAI models at Ignite 🚀 MAI Code 2 hits 98% of GPT‑6.1 Sol's coding quality at ~40% of the price (now in GitHub Copilot), Image 2.7 is #1 on Artificial Analysis image editing, Realtime 1 starts at $0.01/audio min. #AI #Microsoft
2
47
Stop letting AI phrases kill your posts. This free tool rewrites them so they sound like a person wrote them, and it has 70.9K GitHub stars. Humanizer hunts 26 AI-writing tells from Wikipedia's own guide: - "It's not just X, it's Y" - "Let that sink in." - "Let's dive in"
1
102
How to try it: npx skills add blader/humanizer --global --agent '*' then /humanizer + your text. Add a writing sample to match your voice. Catch: needs an agent like Codex or Gemini CLI, and doesn't try to beat AI detectors (per its README). MIT. github.com/blader/humanizer
30
Gemini 4 Neon hit 61.4 at 86 tok/s for $3/$15 per million tokens. It outscored Claude Opus 5.5 (59.1) on coding benchmarks at a 25% lower API cost.
44
🚀 OpenAI drops GPT‑6.2 Astra for heavy‑duty reasoning & coding and GPT‑6.2 Sol for everyday efficiency. Both use fewer tokens, Ultrafast at launch. API pricing: Astra $8 / $40, Sol $2 / $10 per M tokens. #AI #OpenAI
33
Your idle USDT on TRON can now earn up to 7.25% APY from your phone. Aave App just hit Android (early access) with TRC-20 USDT deposits. Skip the waitlist: Ghost Pass code ANDROID, while supplies last. No lock-up. Fine print: variable rate, not a bank, not insured.
2
39
Google DeepMind drops Gemini 4 Neon 🚀 Frontier-level speed for coding, knowledge work & science. #3 on the Artificial Analysis Intelligence Index (61), #2 on Vals. $3/M in, $15/M out, 1M-token context. Live in Gemini API, AI Studio, Antigravity & the Gemini app. #AI #Gemini
57
Claude Sonnet 5.5 hit 70.6% on Terminal-Bench 4.0 at $2/$10. It beat Claude Opus 5.5 (66.4% at $4/$20) on scoped agent tasks while running 30% faster.
2
110
Stop Claude Code from refactoring adjacent files. Save this skill: 1. Create `.claude/skills/surgical-fix/SKILL.md` 2. Add: --- name: surgical-fix description: Use when fixing bugs or updating existing logic. --- Constraints: - Edit only the lines causing the defect. - Never reformat or rename untouched code. - Run `pnpm test` and show passing assertions before closing. 3. Run `/surgical-fix "fix session timeout"` Loads on demand without bloating your base CLAUDE.md.
57
DeepSeek-V3.2 hit 97 tok/s on DeepInfra at $0.29 per million blended tokens. Google Vertex pushed 199.2 tok/s on the same weights at a 2.9x price markup.
1
91
Prevent dirty working trees. Run Claude Code in an isolated worktree: 1. Launch isolated session: `claude --worktree feat-auth` 2. Prompt: "Migrate session tokens to Redis. Run `pnpm test` after edits. If assertions fail, patch the logic in-loop. Commit atomically only when green." 3. Merge when verified: `git merge feat-auth` Keeps your main context untouched while the agent runs risky changes inside `.claude/worktrees/feat-auth/`.
5
1
1
159
Tencent's Hy4 flash just launched. You can use it free in WorkBuddy until Nov 30. Why care: in Tencent's own blind test (85 users, 259 tasks) it nearly tied Claude Haiku 5.5, 2.79 vs 2.81 out of 4, at ¥0.26 a task vs ¥4.73. Best for research, docs, slides. Not long coding.
1
136
Per Tencent's launch post (Nov 16): • 224B params, 12B active per token • 1M-token context, natively multimodal • API: $0.07 in / $0.281 out per 1M tokens, 50% off via Tencent Cloud TokenHub through Nov 30 • Open weights planned, OpenRouter soon hy.tencent.ai/research/hy4-f…
100