VCN #47: Fast Local

Join us for a hands-on build night where bringing a laptop is mandatory! 💻

A local model that's too slow becomes dead weight, and tonight we aim to make it fast. 🚀 In VCN #47: Fast Local, we’ll enhance the performance of your local rig to ensure it can operate efficiently.

Format:

1. The Walkthrough: Discover where the time goes! We’ll cover GGUF quantization tradeoffs, llama.cpp vs vLLM, tensor parallelism, speculative decoding, KV-cache tuning, and how to measure tokens/sec accurately.
2. The Benchmark Bar: We start by running Claude Code powered by z.ai and measuring the hosted speed. It’s essential to know your target for optimizing your local endpoint. 📊
3. Make it Fast: During the hands-on hour, quantize a model, create a fast serving endpoint using Nebius Token Factory credits, tune parameters, and watch your tokens/sec climb! ⏱️
4. Demos: Point your coding agent at your fresh fast endpoint and experience the improvements live.

By the end of the night, you’ll leave with a quick local inference endpoint your coding agent can thrive on. 🌟

Builders Only:

Bring your local rig from Bare Metal or just a model you wish to serve quickly.

🕖 Doors Open: 7 PM
📝 Walkthrough Starts: 7:30 PM
🏢 Location: Frontier Tower, Floor 10

Hosted by: Vibe Coding Nights — Rayyan Zahid (Immersive Commons), Michalis Vasileiadis (Hacker Bob), Eric Mockler (AI Geneticist), Devinder Sodhi (Learning Layer Labs).
Facilitator: Rayyan Zahid. Guest speaker TBD (open call).

👉 RSVP if your local model is private but painfully slow and you want it quick.
Frontier Tower members: Your ticket is complimentary! Please reach out to the team directly for a free RSVP. 🎟️