Simulating local LLM speed before you buy the GPU
Someone built a browser tool that throttles ChatGPT to show you what inference feels like on different hardware configurations.
From the notebook
9 notes on this topic.
Someone built a browser tool that throttles ChatGPT to show you what inference feels like on different hardware configurations.
The most secure Android ROM just said it might not support Google's next flagship phone. The reason is a missing hardware component that makes remote attestation work.
A reverse engineering enthusiast spent four years breaking open the PlayStation 2's Mechacon chip, the hardware guardian that controlled everything from disc authentication to region locking.
The Pixel 11 drops hardware memory tagging that was standard on the Pixel 8 and 9. GrapheneOS called it out on Twitter.
A startup ditched everything GPUs do well and made a chip that runs one architecture 20 times faster.
An LLM reverse-engineered a 2008 HP laser printer's protocol and generated a working macOS driver. No human debugging required.
Needle2 runs on phones, smartwatches, and Raspberry Pis. The entire model is smaller than a single photo.
A firmware update silently killed Transparent Memory Encryption on consumer chips. The engineers stopped replying when users asked why.
An enthusiast loaded a 1T-parameter model into 768GB of Intel Optane DIMMs and got 4 tokens per second on a single GPU. Slow, but it worked.