Blog

notes from the pool

Guides and deep dives from the team building Europe's distributed AI inference network — for GPU owners and for the builders on the other side of the API.

meet mahout: the open-source agent that puts your gpu in the pool

One auditable Rust binary: probe your hardware, enroll with a one-time code, serve verified open-weight models, sign a receipt for every job. How it works, how to install it, and how to contribute.

Read the article →

why we pay gpu owners 80% — and how the math works

Every job leaves an ed25519-signed receipt, the ledger counts in micro-cents, and the 80/20 split follows from who actually pays for the hardware and the power. The full accounting, in the open.

Read the article →

your prompts don't belong in virginia — the case for eu-sovereign inference

Prompts are a live feed of your most sensitive text, and CLOUD Act obligations follow US providers into their EU regions. What Schrems II means for inference, and what our EU-Sovereign tier guarantees.

Read the article →

ship with open-source ai: one api, the best open models of 2026

Qwen3.8, MiniMax M3, GLM-5.2, DeepSeek V4-Flash and Gemma 4 behind one OpenAI-compatible endpoint — swap the base URL, keep your SDK, and pay a transparent per-token price with prepaid spend caps.

Read the article →

put your gaming gpu to work: earn on every token it serves

Your GPU idles twenty hours a day. Three commands put it to work serving open models — with sandboxed engines, checksum-verified weights, auto-pause when you game, and equal NVIDIA and AMD support.

Read the article →