Home / Inference APIs / nafnaf.ai

nafnaf.ai

Free Trial

Swiss GPU marketplace with a 5% platform fee, plus an OpenAI-compatible API with free open models

nafnaf.ai is a GPU marketplace run by Sequester AG in Zug. Independent hosts list GPUs, renters run Docker containers on them, and the hourly rate shown already includes a 5% platform fee. When checked in October 2026 the market listed an RTX 4090 at $0.20/hour and an RTX PRO 6000 Blackwell Max-Q at $0.42/hour, both in Zug.

The same account gets an OpenAI-compatible inference API. Qwen3.8 Flash-Next and DeepSeek V4.1 Flash are free to call; paid models such as MiMo v2.6 Flash draw on a prepaid balance.

Pricing: Hourly

Hosting Cloud
Pricing $0.20 per GPU hour
HQ 🇨🇭 Switzerland
Screenshot of nafnaf.ai webpage

nafnaf.ai Alternatives

Explore 160 products in the Inference APIs category. View all nafnaf.ai alternatives.

CoreWeave icon
Hetzner GPU Servers icon

Hetzner GPU Servers

Dedicated single-GPU servers with NVIDIA RTX PRO Blackwell cards, billed hourly, in Finland and Germany
From From EUR 0.3429/hour, capped at EUR 214/month (GEX45), plus EUR 209 setup, excl. VAT
Airon icon
Eden AI icon

SubQ Snap

Long-context decision model with typed answers and probabilities
Free Trial From $0.04 per 1M input tokens, output free
Bespoke Nimble icon

Bespoke Nimble

Hosted and open decision models that return a probability for every answer
Open Source Free Trial From $0.04 per 1M input tokens, output free

View all Inference APIs tools ≫

Discover more European AI infrastructure ≫

Work on nafnaf.ai? Feature it at the top of Inference APIs.

Is your product missing?

Add it here →