NZ ☄️

First major defeat to Ai in job sector.

🚨 The next AI bill might make you rethink your entire setup. Some developers have been running thousands of AI tasks for months without paying a single dollar in API costs, and it’s not because they found a loophole. One developer built his own local AI setup three months ago and hasn’t looked back. He runs a full AI lab directly from his machine with no API bills, no usage anxiety, and no rate limits watching over his shoulder. The hardware side is two GPUs with a combined 32GB of VRAM, which handles long repetitive workflows and multi-step loops without any concern about token costs. While other developers are refreshing usage dashboards and watching expenses climb, every experiment he runs costs exactly $0. The whole setup runs on two tools: llama.cpp and llama-swap. Every prompt stays on his own device, every test is private, and every month ends without an invoice. But the more interesting shift here isn’t the money saved. It’s what ownership actually feels like when no subscription is deciding how much you can create or experiment. No surprise bills, no throttling, just unlimited runs on your own terms. The cloud isn’t going anywhere, but the people building local rigs today are quietly buying back control over how they work, and positioning themselves well ahead of wherever API pricing goes next. Save this for later.

Hear me out. What if Anthropic just quietly renames Fable 5 to Opus 4.9? The government will never know.

166.09GB… what are you doing ChatGPT? @openai

🚨 Another GPT-5.6 leak 👀 Rumors suggest OpenAI could unveil GPT-5.6 on June 23. > ~3× cheaper than Fable > Up to 1.5M-token context > Stronger agentic coding workflows > Built to compete directly with Claude-style systems The timing is hard to ignore. Many Fable users are expected to move onto premium plans around the same time. If the rumors are true... Does GPT-5.6 steal the spotlight from Fable? 🍿

@i_am_jeganraj not for me, i don’t know what happened haha😂. I worked throughout the day and i still didn’t hit limit.

@CodeByNZ I thought the same. But atleast it reached after few prompts

Am I the only one who noticed that Claude’s rate limits just… aren’t hitting today?

@HixonStudio maybe, i still didn’t hit any limit today but i was on deck throughout the whole day. Productivitymaxxing.

@CodeByNZ Trying to win the customer over during this uknown time of model suspension.

Chinese teams that already distilled Fable 5.

The guy who helped make the future getting locked out of it.

Anthropic: “If we’re going down, everybody’s coming with us.”

Claude Fable 5 unavailable, that was fast.😂

The most unsettling AI take I've read is that it allows wealth to access skill while removing from the skilled the ability to access wealth, and I'm not entirely sure it's wrong.

A few years ago people worried AI would take their jobs. Now governments are worried the wrong people might get access to it.

@EagleCorp @MiniMax_AI What stands out about MiniMax M3 is the flexibility. Pairing that with a 1M-token context window is seriously impressive.

Today, @MiniMax_AI unveiled MiniMax M3, bringing together frontier coding performance, agentic capabilities, native multimodality, and a 1M-token context window in a single open-weights model. We're proud to see MiniMax M3 powered by an open-source speculator built by the @inferact team on top of Hongyang Zhang's EAGLE-3 algorithm from Eagle Corp, helping bring faster, production-grade inference to one of the most capable open models released today. Inferact/MiniMax-M3-EAGLE3 is a 3.3B-parameter draft model trained with TorchSpec’s online speculative decoding framework, combining FSDP training with live vLLM target inference to learn directly from MiniMax-M3 responses and hidden states while preserving the base model’s exact token distribution. Served end-to-end with vLLM, it leverages EAGLE-3’s shared embedding and LM head architecture to deliver efficient, production-grade inference acceleration at scale. Explore the open-source EAGLE-3 speculator powering MiniMax M3 on @huggingface: huggingface.co/Inferact/MiniM…

🚨 Plot twist: The push to shut down Anthropic's most advanced models reportedly came from a rival AI company alleging it could crack Mythos's defenses. Not China.

Sam trying very hard not to smile while reading this announcement.

> be Juan Hernandez > move to the U.S. from Mexico and learn welding for better opportunities > a friend mentions a place called SpaceX > "I have no idea what SpaceX is, but sure" > take the job for $28/hour, not because of the mission > receive $10,000 worth of company stock > spend the next decade welding rockets and buying more shares every paycheck > leave last year and start building rockets for Bezos at Blue Origin > SpaceX goes public at $135 in the biggest IPO in history > your 6,500 shares are suddenly worth more than $1 million He became a millionaire from Musk's rockets while working for Bezos. The only guy who truly won the billionaire space race.

Claude Mythos 5 / Fable 5 is out. > Dominates agentic coding at 80.3% vs GPT 5.5's 58.6% and Gemini's 54.2% > Doubles GPT 5.5 on cybersecurity > Beats every competitor on Humanity's Last Exam Fable mogged every other AI model out there.

@Star_Knight12 nice UI, how many tokens did you burn to get this done ? 😂

built a glass effect today, how's it :)