Sponsored by
🖱
OpenAI just made their best model 14 times faster.
The new "Ultrafast" mode for GPT-5.6 Sol pumps out 750 tokens per second fast enough to power real-time voice agents and live coding assistants without the lag.
But wait, there's more:
ChatGPT can now remember what you did on your Mac yesterday
Google slashed Gemini prices by 50% and released a smarter model in just three weeks
I'm Alex. Welcome to L8R by Innov8.
Let's dive deep 🐰
In today's post:
OpenAI's GPT-5.6 Sol hits 750 tokens per second in new Ultrafast mode
ChatGPT now remembers what you did on your Mac
Google's Gemini 3.7 Flash arrives with 50% price drop
GPT-5.6 Sol Hits Warp Speed at 750 Tokens/Sec
OpenAI just solved the biggest problem with massive AI models.
They teamed up with chipmaker Cerebras to make GPT-5.6 Sol run at crazy speeds without losing any brainpower.
This new Ultrafast mode pumps out words faster than you can blink.
The details:
The new mode hits up to 750 output tokens per second, making it 14 times faster than the standard tier.
Cerebras uses giant wafer-scale chips to keep all the model data in one place, fixing the traffic jams you normally get with GPUs.
OpenAI did not shrink the model to make it fast, so you get the exact same GPT-5.6 Sol smarts.
Companies like Jane Street and Podium are already testing it for real-time stock trading and instant voice bots.
Why it matters:
You used to have to pick between a smart AI that was slow or a dumb AI that was fast.
This update breaks that rule completely.
Now developers can build high-stakes tools, like live financial traders or emergency response bots, using top-tier intelligence.
It proves we do not have to sacrifice speed for smarts anymore.
💡 L8R's Take:
Nvidia should be sweating right now.
Cerebras just proved their giant chips can do things regular GPUs cannot even dream of.
When this rolls out to everyone, it will completely change how we build live AI apps.
From Our Partner
Wake Up Smarter About AI.
The president of OpenAI. The CEO of Google. The founder of LinkedIn. We talk to the people building AI. Every morning, we put what they told us in your inbox.
What they're actually worried about. What they're betting on. What's coming next. Delivered in five minutes, before your first meeting.
With one email, and only five minutes, you can stay ahead of 99% of the world.
ChatGPT Now Remembers Everything You Do on Your Mac
OpenAI just gave ChatGPT a massive memory upgrade for your Mac.
The new Computer History feature tracks your clicks and typing to understand exactly how you work.
It launched on August 13, 2026, and it completely changes how AI helps you get things done.
The details:
It uses the macOS accessibility API to track your clicks, typing, and app switches instead of taking screenshots.
Your data stays on your device and turns into local memory files for ChatGPT and Codex to read.
The AI spots repetitive tasks you do and offers to build custom Skills to automate them.
It is strictly opt-in, meaning you must turn it on yourself, and it ignores private browsing.
Right now, it is only for Pro, Business, and Enterprise users on macOS.
Why it matters:
You no longer have to explain your whole project every time you open ChatGPT.
The AI already knows what apps you used and what you typed. It turns your daily grind into automated shortcuts.
This pushes AI from a simple chatbot to an invisible assistant that lives in your desktop.
💡 L8R's Take:
This is the exact feature we need to make AI actually useful for boring office work.
People will freak out about privacy, but keeping the data local is a brilliant move by OpenAI.
Turn it on, let it watch your boring tasks, and let it do the heavy lifting for you.
From Our Partners
Short on cash? Get up to $750, no credit check.
Is payday still a few days away, but the bills aren't waiting? We've all been there. That's why there's Klover — a cash advance of up to $750* from your upcoming paycheck, with no credit check and no late fees. Ever.
Cash in 3 Easy Steps
Sign up — Creating your Klover account takes just minutes.
Link your bank — Securely connect the account where your paycheck lands.
Get your cash — Access up to $750* within minutes for a small fee, or wait 3 days and get it for free.
No credit check. No late fees. No catch — just a little breathing room until payday.
*Not all users will qualify. Advances range from $25–$750; average advance is $170. Express transfer fees may apply.
Google Drops Gemini 3.7 Flash and Slashes Prices by 50%
Google just dropped Gemini 3.7 Flash out of nowhere.
This new model is built to write code and run AI agents, and they cut the price in half to get you to use it.
It is an aggressive move to win over developers right now.
The details:
Google released Gemini 3.7 Flash on August 13, 2026, just three weeks after the 3.6 version.
The model scores 65.3% on the DeepSWE coding test, which is a huge jump from 49% on the last version.
Prices are cut in half through the end of 2026, dropping costs to $0.75 per million input tokens and $3.75 per million output tokens.'
It features a 1-million token context window and lets you choose between three different thinking levels.
This launch happens right as Koray Kavukcuoglu takes over Google DeepMind from Demis Hassabis.
Why it matters:
Google wants you to build AI agents on their platform today.
By dropping the price and boosting coding skills, they are making it super cheap to test huge automated workflows.
They know developers hate when AI makes up fake answers, so this model focuses on fixing errors and staying on track.
💡 L8R's Take:
Athree-week update cycle is totally insane.Google is dropping cheap, fast models to grab your attention while we wait for their big Pro model.
Take advantage of this crazy discount while you can, because the cost doubles on January 1.
🚀 Quick L8R Summary
OpenAI Ultrafast: GPT-5.6 Sol now hits 750 tokens/sec with new Cerebras-powered mode 14x faster than standard tier.
Computer History: ChatGPT on Mac can now peek at your past screen activity to give smarter answers and automation ideas.
Gemini 3.7 Flash: Google's newest coding model drops with 50% cheaper pricing and better debugging chops.
📩 Innathe L8R engane undarunnu 👇?
We read every reply - just reply to this email and let us know how we can improve !
Appo adutha L8R il kanaam bie…👋