Two bars of the same 15.6 GB of memory. Before: the agent takes 9.4 GB while rendering a video, leaving 6.2 GB for everything else and causing a 15 minute freeze. After: a hard fence caps the agent at 6 GB, Unraid keeps 1.7 GB, 1.2 GB is protected for DNS, media and chat, and 6.7 GB is shared by everyone else.

Sixteen gigabytes and an agent that wanted ten

I should have bought RAM last year. Anyone who runs anything at home has said some version of that sentence in 2026. Conventional DRAM contract prices rose 93 to 98 percent in the first quarter of this year alone, according to TrendForce, and they are still climbing, with another 10 to 15 percent forecast for this quarter (TrendForce, 30 September 2026). Graphics cards are worse. TweakTown found the cheapest RTX 5090 in Germany listed at about $5,900 (₹5,61,000), against a list price of about $2,360 (₹2,24,000), converted at 18 October 2026 rates. The AI build-out is buying the memory, and the rest of us get what is left. ...

October 18, 2026 · 8 min · Romit Raj
The LLM cost proxy dashboard: daily token bars and a per-model cost table.

Metering what my Hermes agent spends

My agent, Hermes, talks to me on a Matrix server I run myself. Self-hosting the chat window was a privacy decision before it was anything else, and that setup is a post for another day. What matters here: an agent that chats all day needs a meter. So every LLM call it makes goes through a proxy I wrote — and “wrote” deserves an asterisk, because I’m a lawyer. An LLM helps me code things out; I mostly argue with it until it works. One SQLite file tells me what it spent today. ...

September 20, 2026 · 2 min · Romit Raj