Put a Display Showing Your Token Usage in Your Face, Says Local Inference User
natesiggard · x · 2026-10-18
A quick practical tip: once you run local inference, set up a display showing your token usage somewhere you'll see it every day, so it nudges you when usage starts leveling off — a behavioral feedback loop for your AI habits.
More from Infra
- Rust-based Git server on S3 handles 7,600+ concurrent pushes per second — blaizedsouza · 2026-10-18
- Chaining 2 PCs, a MacBook and 2 iPhones Runs 48-Expert MoE at 37 tok/s on 60k Context — StayLameBro · 2026-10-18
- Ollama US hosting coming soon, bringing DeepSeek-V4.1-Flash-1017 stateside — gaganghotra_ · 2026-10-18
- 43-minute video encoded to AV1 in 2m35s on Intel Arc A380, 77% smaller — haydendevs · 2026-10-18
- NYT: CoreWeave's financial arrangements likened to WeWork as AI debt worries grow — SumitGup · 2026-10-18
- 32GB→64GB RAM lets a single RTX 4090 run Qwen 4-bit quants at 90+ tok/s — Distinct-Pie2389 · 2026-10-18