Live · OpenAI-compatible

Frontier models on hardware you can trust

Fast LLM inference served from our own NVIDIA DGX supercomputers. Your prompts never touch a third-party cloud. Prepaid credits, no subscription, no seat fees — drop in one base URL and go.

your first request
# works with any OpenAI SDK — just change the base URL
curl https://monkey.reinforce42.com/v1/chat/completions \
  -H "Authorization: Bearer sk-your-key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-oss-120b",
    "messages": [{"role":"user","content":"Explain MoE in one line."}]
  }'
Runs on 2× NVIDIA DGX Spark (GB10)
Your data never leaves our hardware
OpenAI-compatible — zero code rewrite
Pay only for tokens you use
The lineup

Open frontier models, metered per token

Prices are per 1M tokens (input / output). Credits are drawn down exactly as you use them — no rounding, no minimums.

gpt-oss-120b
OpenAI's 120B open reasoning model. Best general quality & tool use.
$0.09 / $0.45 per 1M
deepseek-r1-70b
Deep step-by-step reasoning for math, logic and code.
$0.10 / $0.40 per 1M
gemma3-27b
Fast, lightweight, great for high-volume everyday tasks.
$0.05 / $0.20 per 1M
Prepaid credits

Buy a pack, get a key in seconds

Pay once, receive an API key preloaded with credit. Bigger packs include bonus credit. Top up any time — keys never expire.

Starter
$10
$10 in API credit
  • All models included
  • 60 requests / min
  • Key never expires
  • Usage dashboard
BEST VALUE
Builder
$25
$27 in API credit · +8% bonus
  • Everything in Starter
  • 120 requests / min
  • Priority routing
  • Email support
Scale
$100
$115 in API credit · +15% bonus
  • Everything in Builder
  • 300 requests / min
  • Highest throughput
  • Direct support line

Secure checkout via Stripe · cards & wallets · instant key delivery

Quickstart

From zero to first token in three steps

Buy credits

Pick a pack above and pay with Stripe. Your API key appears instantly on the confirmation screen.

Point your SDK

Set the base URL to monkey.reinforce42.com/v1 and drop in your key. Any OpenAI client works unchanged.

Call any model

Send chat completions to gpt-oss-120b and friends. Credit is drawn down per token; track it live in your dashboard.