NexusComputeBETA
Back to Endpoints
Cold — first request slower
gpt-4o

Nova A100 GPT-4o Endpoint

GPT-4o · Nova A100 Cluster · EU-West

$0.1060
per 1k tokens
Status
Cold
Cold Start
~4.8s
Total Calls
3
Tokens Served
0.3k
Quick Start
curl
curl -X POST /api/endpoints/1/infer \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{"prompt": "Hello, world!", "maxTokens": 256}'

Endpoint URL: /api/endpoints/1/infer

Test Console
Hardware
GPU
NVIDIA A100 80GB
Count
8×
VRAM
640 GB
Uptime
99.95%
Location
EU-West
Provider
Nova A100 Cluster
Model
Model
GPT-4o
Provider
OpenAI
Context
128,000 tokens
Provider Rate$0.0060/1k
Platform Fee$0.1000/1k
Total$0.1060/1k
Jobs are queued and dispatched to the provider's agent. Cold starts add ~4.8s; warm endpoints respond faster. The test console polls until your job completes.