I gave 4 "Flash" models the same job: redact the PII in an HR letter using an agentic mask tool.
One of them cost 630× more than another. And it wasn't 630× better. Results 🧵 🥇 GLM 5.3 Flash $0.005, 15 tool calls Clean sweep. Overshot a bit (also redacted employee ID +
I made an AI inbox plug-in (AI Inbox) to solve the problem that I have many AI conversations and forget to check the replies.
AI Inbox currently monitors the responses to ChatGPT, Deepseek and Claude web pages in the browser (generating or response completed)
This conspiracy is way too good to ignore
Moon dev: "deepseek is getting absolutely crushed right now and nobody is talking about why" "here's the part that broke my brain, didn't the hedge fund build this thing themselves"
Yesterday ChatGPT, Claude, and Grok all went dark at the same time.
This is how I rigged the home lab, so mine doesn’t. 🧠 DGX Spark — DeepSeek-V4-Flash-0731, 384K (reasoning) 👁 Spare PC RTX 3060 12GB — Qwen3-VL-8B (vision) 🎨 RTX 5080 PC — Pinokio + ComfyUI (image / video /
You have reached the end of the archive
All of deepseek