Ai & Blockchain lover

Only live updates Ai & Blockchain

[FN @FN_based]

Been testing an AI model for deep research on my local grocery business over the past few days. Honestly? Impressed.

The reasoning is solid, and the actual data (links, prices) is 100% accurate. Zero hallucinations, just facts. A genuine discovery.

[Ai & Blockchain lover @frndlyrefr]

I've had perplexity for 3 years at least and @Apodex_AI revolutionizing so I switched over. When will there be an app for it?

[Apodex @Apodex_AI]

Meet 𝗔𝗽𝗼𝗱𝗲𝘅 𝟭.𝟬 🔭 — a heavy-duty agent team for deep research, which sets the SOTA! The team searches the web, reasons over evidence, and writes reports where every claim is backed by an explicit 𝘦𝘷𝘪𝘥𝘦𝘯𝘤𝘦 𝘤𝘩𝘢𝘪𝘯, independently audited before delivery.
🌐 https://t.co/pOQAjL92uF

[Apodex @Apodex_AI]

Meet 𝗔𝗽𝗼𝗱𝗲𝘅 𝟭.𝟬 🔭 — a heavy-duty agent team for deep research, which sets the SOTA! The team searches the web, reasons over evidence, and writes reports where every claim is backed by an explicit 𝘦𝘷𝘪𝘥𝘦𝘯𝘤𝘦 𝘤𝘩𝘢𝘪𝘯, independently audited before delivery.
🌐 https://t.co/pOQAjL92uF

[Nicolas Camara @nickscamara_]

We just made Firecrawl free again!

First open source, then 1,000 free credits in our cloud and now you don’t even need to sign up.

Your agent deserves the best web data available

[Ai & Blockchain lover @frndlyrefr]

🚀 Just discovered https://t.co/k6ONuURG4R, an amazing AI creation platform!
✨ Unleash your creativity and generate stunning images and videos.
Join me on this creative journey!
👉 https://t.co/OKnaXTFNQr

#AI #Art #HotgenAI

[MIKE @mikenevermiss]

AMD JUST KILLED AI SUBSCRIPTIONS

AMD CEO Lisa Su officially unveiled a lunchbox-sized PC and ran a live demo of a 235 billion parameter model

no data center. no cloud. no rented GPUs

the chip inside is the AMD Ryzen AI Max+ 395

it’s the first x86 chip where the CPU and GPU share the same memory block

up to 128 GB of unified memory

an RTX 5090 gives you 32 GB of video memory

a 4090 gives you 24 GB

but this little machine offers more than three times the memory of either of them

and it fits in a backpack

in inference with DeepSeek R1, it beat an RTX 5080 by 3x

a desktop the size of a thick book outperforming a graphics card worth over a thousand dollars in a real AI workload

now do the math on your subscriptions

Claude Code Max: $200 per month
ChatGPT Pro: $200
Cursor: $20
Gemini: $20

that’s $5,280 a year before building a single thing

the 128GB version of this machine costs between $1,800 and $2,500

at that rate, it pays for itself in less than a year

and then it runs with no additional costs, FREE

Install Ollama
Download Qwen3 235B
Point Claude Code to localhost
The same interface you’re already using
Nothing leaves your machine
Nothing costs per request
No limitations at 3am when you finally have time to build

lawyers stop worrying about what OpenAI does with their files

developers stop watching the token counter

founders stop killing prototypes because the cloud bill scares them

local AI is no longer just a cheaper option

it’s the only AI that no one can take away from you

and the question is no longer whether local AI is good enough

it’s clear that it is

the real question is why keep paying subscriptions every month when you can run it yourself

[Thanh Nguyen @ng_thanh8]

Claude Opus 4.8, Sonnet 4.6, and Haiku are all available for free right now.

The offer includes more than $350 in credits, plus an $80 signup bonus that expires in about 24 hours.

After that, users can continue with $10 every 5 hours and $70 per week for a full month.

No card required.
https://t.co/9bmSuFjI30

[Jiya Ziyue @JZiyue_]

Free 1 minggu di ZenMux — GLM-5.2 live 🔥

Specs:
→ 1M context
→ Coding + agentic improvements
→ #2 global Code Arena
→ Same price as 5.1

Free window: 1 week. Claim sekarang 👀

https://t.co/AQiXHRv63Q

[Ai & Blockchain lover @frndlyrefr]

https://t.co/OKnaXTFNQr

[Pankaj Kumar @pankajkumar_dev]

Gemini 3.5 Pro Leaks: Launch Is Getting Closer

- Google has started hinting at Gemini 3.5 Pro, with a "3.5 Pro coming soon" tag appearing on the Gemini 3.1 Pro model card.
- Expect stronger vision, better multimodal reasoning, and improved SVG/frontend generation over Gemini 3.1 Pro.
- It will likely ship with stricter safety filters and content moderation.
- Pricing is expected to be higher than Gemini 3.1 Pro.
- The biggest hope: Google fixes the laziness on long and complex tasks seen in earlier 3.5 Pro checkpoints before the official launch.

[Atenov int. @Atenov_D]

FREE 1 month Plus subscription. Seedance 2.0, Kling 3.0, Veo 3.1, WAN. No card, no phone. Just 6 tasks in your dashboard.

The platform is HotGen AI. Top video and image generation models in one place, normally locked behind a paid tier.

Here's the exact flow:

→ Go to hotgen(.)ai
→ Sign In / Sign Up with Google - no card, no phone required
→ Open the Tasks section in your dashboard
→ Complete all 6 tasks - generate an image, generate a video, share, a few others
→ 1 month of Plus activates automatically

Plus unlocks higher generation limits across Seedance 2.0, Kling 3.0, Google Veo 3.1, and WAN.

10 minutes of clicking through tasks for a month of access that normally costs real money.

Bookmark this before you lose the link.

[J A Z I I @notjazii]

GLM 5.2 vs Kimi K2.7

both were tested on same settings with same prompt

> GLM one shotted everything in 35 mins
> Kimi needed extra prompts to fix movements and took 30 mins

super surpised by how good GLM 5.2 is, while being so cheap

which one do you think won?

[Thanh Nguyen @ng_thanh8]

Claude Fable can currently be accessed for free through GitLab’s 1-month trial.

Just sign up with Gmail, complete the onboarding tasks, and open the AI chat icon in the bottom-left corner.

No KYC or card needed, but the free credits are very limited — even a “hi” can cost 1+ credit, and you only get 24 credits total.

[Higgsfield AI 🧩 @higgsfield]

Scriptwriting with Claude Fable 5 + Higgsfield MCP is on another level.

Both videos made using Seedance 2.0.

[Russell Brunson @russellbrunson]

You can’t sell to everyone. Pick your niche and go deep.

[Ai & Blockchain lover @frndlyrefr]

#Claude5 #Mythos tomorrow

[Dan McAteer @daniel_mac8]

Merry Claude Mythos eve to all who celebrate.

[Ai & Blockchain lover @frndlyrefr]

#Claude #mythos very soon

[Alex Heath @alexeheath]

Sources: Anthropic is planning to release a public version of Mythos tomorrow
- Will have substantial guardrails and not be as cyber permissive as what Project Glasswing partners can access
- Will be dramatically better at long-horizon, multi-turn tasks https://t.co/cqq3Y5CVb0

[Hridoy Reh @hridoyreh]

How to find business ideas:

1. Go to Flippa.
2. Filter websites doing $10,000+/m.
3. Analyze the website / SaaS now.
4. Pick a product idea that has demand.
5. Build with better features.

Do SEO and make money...

[Sharbel @sharbel]

Someone built an AI agent that searches Reddit, X, YouTube, HN, TikTok, Polymarket, and the web in parallel.

Scores everything by real upvotes, real likes, and real money.

Synthesizes it into one brief. In seconds.

It's called /last30days. 28,700+ stars on GitHub.

You type one command. The agent fans out across every platform at once. Reddit threads. X posts. YouTube transcripts. Polymarket odds backed by actual money. HN comments. GitHub commits. It scores each source by what real people engaged with. An AI judge synthesizes the whole thing into one grounded summary of the last 30 days.

Here's what it does:

→ Searches Reddit for top upvoted threads and comments on any topic, person, or company.
→ Pulls X posts and scores them by likes and recency. Not algorithmic feed. Raw signal.
→ Transcribes and searches YouTube videos. Finds what was actually said, not just the title.
→ Reads TikTok engagement. Surfaces what creators and communities are actually talking about.
→ Queries Polymarket odds. Real money bet by real people on what happens next.
→ Searches Hacker News. The technical community's unfiltered take.
→ Searches GitHub commits and PRs. What someone is actually shipping right now.
→ Runs all sources in parallel. Scores them against each other by engagement weight.
→ AI agent judge synthesizes everything into one brief. No raw dump. A grounded summary.
→ Zero config to start. Reddit, HN, Polymarket, and GitHub work immediately.
→ One setup wizard unlocks X, YouTube, TikTok, and more in 30 seconds.
→ Installs into Claude Code, Codex, Cursor, Copilot, Gemini CLI, and 50+ agent hosts.

Here's the wildest part:

Google doesn't touch Reddit comments or X posts. ChatGPT has a Reddit deal but can't search X or TikTok. Gemini has YouTube but not Reddit. Claude has none of them natively. Every platform is a walled garden with its own API, its own tokens, its own auth.

No single AI has access to all of it. Until you bring your own keys and bridge them with an agent.

That's the unlock. Not one better search engine. A dozen disconnected platforms, scored against each other by what real people actually engaged with and bet real money on.

Google aggregates editors. /last30days searches people.

Perplexity Pro: $20/month. $240/year.
ChatGPT Plus: $20/month. $240/year.
You(dot)com Pro: $15/month. $180/year.

/last30days: $0. Unlimited queries. Unlimited topics. Your API keys. Your agent. Forever.

28,700+ stars. 2,431 forks. MIT licensed.

MIT licensed. Self-hosted. Open protocol. Free forever.

100% Open Source.

Github repo: https://t.co/jSgEaiye2H

[Ai & Blockchain lover @frndlyrefr]

#Claude5 #mythos very soon

[Dan McAteer @daniel_mac8]

Claude Mythos 5 likely lands this week. Either it leapfrogs all rivals and Anthropic pulls ahead in the AGI race, or it's expensive hype.

My bet: it leapfrogs.

It broke METR's benchmark and solved AISI cyber challenges no model has.

Mythos is the inflection point.