Loading…
Topic
Sprite
3 posts about Sprite. Every summary links to the original.
- 3 posts
- 1 company
The Good Stuff
Paid picks we use ourselves. Links carry a referral code — a subscription through one earns us a commission at no extra cost to you.
Newest first
Posts
Run Paperclip AI on a Sprite
Paperclip is an open-source orchestration platform for creating teams of AI agents that operate like a company with goals and autonomous execution. This guide shows how to run it on a Sprite, a hardware-isolated Linux environment with 100GB storage, a TLS URL, and a pause/resume lifecycle that charges only for compute in use. Setup requires creating a Sprite, running npx paperclipai onboard -y to accept defaults like embedded PostgreSQL and local storage, then editing the config to bind to 0.0.0.0 in authenticated mode. The Sprite’s hostname must be added to Paperclip’s allowlist with npx paperclipai allowed-hostname to avoid an error, and the service registered via sprite-env services create before accessing the dashboard through the Sprite’s URL. Sprites provide checkpoints for rollback, automatic pausing when idle, hardware isolation to contain agent actions, and network policy to restrict HTTP access to allowed domains.
Run OpenClaw (Formerly Moltbot and Clawdbot) in a Sprite
OpenClaw is an open-source AI assistant that connects to messaging platforms like WhatsApp and Telegram. Running it with generous permissions is risky, so the post describes deploying it inside a Sprite, a hardware-isolated Linux computer that sleeps when idle and wakes on request. The tutorial walks through creating a Sprite using the Sprites CLI, installing OpenClaw with its install script, registering the Gateway as a Sprite service that auto-starts on HTTP requests, and approving device pairing from the console. After pairing, users access the dashboard via the Sprite URL and token, where they can start chatting with OpenClaw and ask it to set up integrations. As a bonus, the post instructs OpenClaw to create a Sprite skill by reading documentation files, enabling it to manage services, checkpoints, and network policy.
Run a Local LLM on a CPU With BitNet and a Sprite
Microsoft’s BitNet uses 1.58-bit quantization and ternary weights to run a 2.4 billion parameter model on a CPU, fitting in 1.1GB of RAM and achieving 50 tokens per second. The author deploys this model on a Sprite, a hardware-isolated compute unit that provides a URL with TLS, sleeps when idle, and wakes on request with a 1.2-second model load time. A Python HTTP server wraps llama-cli as a subprocess, exposing an OpenAI-compatible API at /v1/chat/completions, a status page, and a self-addressed client at /client.py. Checkpoints capture the entire environment atomically, allowing safe rollback when experimenting with different models, kernel parameters, or prompt formats. The Sprite registers as a service on port 8080, making the inference endpoint available to any other Sprite in the account without infrastructure changes.
Related topics
All topics →Weekly digest
The best engineering reads, once a week.