Welcome, humans.

There’s only 3 days left of The Neuron this year, and our end of year subscriber drive is still in full swing! For every person you share The Neuron with, you get one entry to win free AI for a full year!

You share with one person? You get one entry. You share with 50 people? You get fifty entries. Granted, these have to be real people…

Share The Neuron

You currently have 0 referrals, only 2 away from receiving PLACEHOLDER MILESTONE NAME.

Click to Share

Or copy and paste this link to others: https://www.theneurondaily.com/subscribe?ref=PLACEHOLDER_CODE

Here’s what you need to know about AI today:

• Devs can now build with o1—we break down what that means for you.
• Sakana AI cuts LLM memory costs by 75%.
• NVIDIA released $250 mini supercomputer for AI.
• Salesforce to hire 2K reps for new AI tool.

Advertise in The Neuron here.

OpenAI and NVIDIA are making AI more accessible than ever.

Yesterday, OpenAI announced a massive developer update that includes better models, new tools, and (finally!) lower prices. And NVIDIA had an interesting surprise, too.

First up: OpenAI o1 is now available to developers. This is their most advanced reasoning model yet—and it comes with some fancy new features:

• It can “see” and analyze images (great for science and manufacturing).
• It has something called “reasoning effort” control (basically, you can tell it how long to think before answering).
• It can generate super-specific outputs (perfect for businesses that need structured data).

Fun fact: o1 is so good at math, it scored better on the American Invitational Mathematics Examination than most human math whizzes. Now if only it could help us with our taxes...

But the big news is how OpenAI has lowered prices for services that matter: The new o1 uses 60% fewer "reasoning tokens" per request, but still costs about $15 per million (vs $2.50 for GPT-4o) for input and $60 per million (vs $10 for GPT-4o).

There's also a budget-friendly o1-mini at $3 for input and $12 for output tokens per million. And GPT 4o with voice is also cheaper now:

1. Voice input: 60% cheaper at $40 per million tokens.
2. Voice output: $80 per million tokens.
3. Cached audio: 87.5% cheaper at $2.50 per million tokens.

GPT-4o mini offers similar features at roughly 1/10th the original cost—think Netflix's tiered pricing model; you're getting most of the same features, just slightly lower quality.

Voice AI has been expensive to run, which is why we haven't seen many voice-based AI apps hit the market yet. These price cuts could change that—imagine more companies being able to afford AI customer service calls or voice-based tutoring systems.

The coolest price drop of all, though? NVIDIA’s new $250 AI supercomputer.

NVIDIA released the new Jetson Orin Nano Super a palm-sized supercomputer chip for running local AI models (like Llama 8B) and robotics projects, like:

• Visual AI agents (think smart cameras that understand what they're seeing).
• Robotics projects (from robot arms to autonomous rovers).
• Computer vision applications (like smart retail systems or industrial inspection).

Check out this launch video… the ending is super goofy.

Introducing NVIDIA Jetson Orin™ Nano Super: The World’s Most Affordable Generative AI Computer

Why this matters: Previously, experimenting with genAI required expensive hardware or cloud credits. NVIDIA's device offers an affordable alternative—like a mini AI data center for your desk, perfect for running models through tools LM Studio or Ollama.

These developments make advanced AI more accessible for businesses and users alike. Cheaper AI = more companies using it = more AI tools for everyone!

FROM OUR PARTNERS

This Disruptor is a Modern-Day Robin Hood

Take from the rich, give to the people. This made Robin Hood a hero. Well, with companies like Facebook making $42B a year off user data, Mode Mobile is the hero we need today.

Mode pays customers for simply using their phone. That’s something people already do 30 hours a week on average. No wonder their customers have earned and saved over $325M, making Mode over $60M in revenue.

That earned Mode the top spot on Deloitte’s most recent list of fastest-growing software companies. They even recently reserved the Nasdaq ticker $MODE.

The deadline to Invest and receive your shares this year is 12/20. Invest at $0.26/share today.*

Please read the offering circular at invest.modemobile.com. This is a paid advertisement for Mode Mobile’s Regulation A Offering. A reservation of the ticker symbol is not a guarantee that we will be listed on the NASDAQ. Any IPO timing is unknown, general steps to be accepted have not been undertaken at this point, and that listing is not guaranteed.

Treats To Try.

1. *Transform your video content with AI. See how Dell’s Precision workstations with NVIDIA RTX™ power Nuke CopyCat and Beeble.ai. Register for Jan 9th!
2. Nora offers you 24/7 anonymous mental health support when you need someone to talk to (iOS, Google Play)—free option + paid tiers.
3. Findr is your personal search engine that remembers everything for you across apps, files, and notes (five free uploads / searches per day).
4. Wondercraft creates complete podcast episodes from text prompts by generating scripts and combining AI voices with music on a timeline editor.
5. OpenCord.AI automatically replies to your Reddit + X comments and turns them into sales opportunities by engaging with potential customers.
6. Aimfox automates your LinkedIn outreach—it manages connections through personalized messaging and performance from one dashboard (free trial avail).
7. Newsletter Name Generator helps you create catchy names for your newsletter in 15 seconds.

See our top 51 AI Tools for Business here!

*This is sponsored content. Advertise in The Neuron here.

Around the Horn.

• Salesforce announced it will hire 2K reps to sell the 2nd generation of its Agentforce AI tool starting in 2025.
• China is creating really popular language models and keeping pace with the U.S. thanks to Alibaba’s Qwen series and DeepSeek, which replicates o1’s reasoning methodology—there’s also Talkie from MiniMax, who’s giving Character.AI a run for its money as the 12th most downloaded AI app.
• New data from cap table management company Carta found non-AI startups are having a harder time raising money at high valuations than AI ones.

FROM OUR PARTNERS

Stop Robocalls, Spam Emails, Stalking, & More…

Americans lost $39.5B to phone scams last year, and major data breaches exposed millions of numbers. Surfshark's Alternative Number gives you a virtual US phone number for shopping, dating apps, and travel bookings—keeping your real number private and spam-free. Stay protected here.

Under the Hood

1. Hermes 3 lets you fine-tune how your AI assistant behaves without special prompting - it processes your natural requests with clear reasoning and works with external tools like data searches and code.
2. Command R7B helps you chat, code, answer questions and work with external data sources in 23 languages while taking up less computing power than bigger models.
3. Sakana AI developed a technique called CycleQD that reduces LLM memory usage by up to 75% using “neural attention memory modules” (paper here).
4. Prompt Wizard from Microsoft automates the creation and refinement of prompts for language models through a feedback-driven process that learns from its own improvements (code here).

#Where do you Neuron?!

Submit where you #Neuron here for a chance to be featured in our newsletter next week! Rules to get featured:

• Monitors / phones in a unique location.
• If you include your face, we’ll protect your identity (see below).
• Cats = heavily encouraged.
• Dogs = case by case basis.

Kim from New Rochelle, NY: Eddie and I so enjoy checking in with the Neuron and getting our hit of catnip every morning!

Mavka from Kyiv (one of our sponsors’ cats!): “I may be a cat, but I’ve learned to nap through air raid sirens, blackouts, and all the noise. Meow from Ukraine!”

A Cat's Commentary.