Stopdepending on big AI

Open models. Lower costs.

TRY CHAT

AI should be accessible,
cheap and free from corporate oversight

Low prices

Open models

Zero data retention

Calculate how much you save

Daily volume (input + output)10M tokens
1M100M

$0.2

your price

$40

corporate average

You save

$40

99% lower cost

First 100 million tokens free

try it

DEMO CHAT — try it

Connectto the model,

that fits your task

curl https://inference.dahl.global/v1/chat/completions \
  -H "Authorization: Bearer ..." \
  -H "Content-Type: application/json" \
  -d '{
      "model": "MiniMaxAI/MiniMax-M2.7",
      "messages": [{"role": "user", "content": "Hello!"}]
  }'

Your API key

Our approach

Affordable AI inference

Get access to powerful open models through a decentralized GPU network built to reduce inference costs for developers, startups, and AI teams.

Independence from providers

Connect to specialized models ready for commercial use without being locked into a cloud giant, closed ecosystem, or proprietary AI stack.

Built for real tasks

Use a simple API to test, compare, and launch models for code generation, chats, autonomous agents, and AI apps with a fast start and transparent terms.

Fresh news

read

Frequently asked questions