Free voice cloning now runs inside a browser tab. No install, no server, no monthly plan.
Zyphra just open-sourced ZONOS3-Mini (Apache 2.0):
- clone a voice from an audio sample
- set accent, age, emotion, pace
- steer delivery word by word
- runs on your own GPU via WebGPU
1
21
Most DTC perfume ads look like generic stock footage.
We built this launch film at Goldgrain Studio. Get 5 hook-tested video ads for $490.
1
17
Stop paying for crypto trading bots.
This one is free, open source and has 56.7K GitHub stars.
Freqtrade:
- backtest any strategy on real exchange data
- dry-run it with zero money in
- control it from Telegram or a web UI
- spot + futures
An AI agent now chases LP fees on tokenized stocks on Robinhood Chain. The clever part is when it quits.
Circuit's first agent:
- funds the top Stock Token/USDG pools on Metric
- checks them every minute
- steps out at the open, jobs, CPI, Fed and earnings nights
No leverage.
Fine print, per the official docs:
→ unclaimed links expire ~33h after the last publish; claim to keep them
→ Personal plan: $0, no card, $1/mo of usage, custom domain, database
→ new spaces start private, so make a viewer link before sharing
spacefast.com/docs/cli
7
How to put anything your AI built online for $0, no account (3 steps):
1. Let ChatGPT, Cursor or any AI save the site to a folder
2. Run: npx spacefast publish
3. Get a live link + a private claim link
Launched yesterday by WordPress's parent, Automattic. Save this.
How to stop Cursor, Codex or Claude Code writing outdated Gemini API code (3 steps):
1. Google's official skill:
npx skills add google-gemini/gemini-skills --skill gemini-api-dev --global
2. Live docs MCP:
npx add-mcp "gemini-api-docs-mcp.dev"
3. Restart your editor/terminal.
Running coding agents this week? The 3-repo stack that keeps them from wrecking your repo:
1. mattpocock/skills → agent grills you first
2. stablyai/orca → parallel agents, one worktree each
3. cloudflare/security-audit-skill → audit before merge
Setup below. Save this.
SGLang hits 16,200 tokens/sec on an H100 vs vLLM's 12,500 running Llama 3.1 8B.
RadixAttention prefix caching delivers 29% more throughput on identical silicon.
1
22
Meta just shipped Muse Ember.
Best use isn't the benchmark. It's the split:
→ big model plans
→ Ember does the repeat work: tool calls, tests, ticket sorting
$0.10/M in, $0.50/M out. Same request body as Muse Spark.
Catch: Haiku 5.5 still wins on mergeable code. Save this.
Claude just gave paid users up to 3 FREE limit resets. Most people will never find the button.
Claude Fable 5.2 dropped yesterday. Here's what actually changed:
→ Free resets: Max/Ultra get 1 to 3, based on how long you've subscribed (under 6 months = 1, 6-12 months = 2, over a year = 3). Pro gets 1.
→ Where: Settings → Usage → Limit resets → "Reset for free"
→ Speed: up to 30% faster than 5.1
→ Cost: Fable on High matched Opus 5.5 on Max on a hard coding test, for roughly half the cost per task
→ API: 20% cheaper, $8 in / $40 out per 1M tokens
The catch: paid plans only. Pro's free week ends Nov 16, so use it now.
Pro move: Fable for big code and deep research. Opus 5.5 for everyday chat.
Watch the 60-second breakdown above, and subscribe to Decoded AI on YouTube. We break down every AI drop like this the day it lands. Link in the reply.
ChatGPT can now turn your dog into a GIF. Free.
Images 2.6 saves your pet's face once. Mention it in any new chat — same dog, zero re-uploads. Animate it. Fix one broken frame without redoing the whole GIF.
youtube.com/shorts/yFgdx7o13…
Subscribe to Decoded AI on YouTube













