EBS AI API

OpenAI-compatible endpoint on EBSsystems.co hardware. Base URL https://api.ebssystems.co/v1. Your key is on your service page in the client area.

Models

model	what it is	context
`qwen2.5-32b`	Qwen 2.5 32B Instruct	8192
`qwen3-30b-a3b`	Qwen3 30B-A3B (fast MoE)	8192
`qwen3-8b-members`	Qwen3 8B Community (old name)	4096
`qwen3-8b:community`	Qwen3 8B Community (member GPUs, EBS GPUs when none is free; not for sensitive data)	4096
`nomic-embed-text`	Nomic Embed Text v1.5	8192

curl

curl https://api.ebssystems.co/v1/chat/completions \
  -H "Authorization: Bearer ebs-..." -H "Content-Type: application/json" \
  -d '{"model":"qwen2.5-32b","messages":[{"role":"user","content":"Hello"}]}'

Python (openai package)

from openai import OpenAI
client = OpenAI(base_url="https://api.ebssystems.co/v1", api_key="ebs-...")
r = client.chat.completions.create(model="qwen2.5-32b", messages=[{"role": "user", "content": "Hello"}])
print(r.choices[0].message.content)

Endpoints

POST /v1/chat/completions (streaming supported) · POST /v1/completions · POST /v1/embeddings · GET /v1/models. Limits per plan: requests per minute, tokens per day, concurrent requests; a 429 tells you which. Prompts are never stored. Questions: support@ebssystems.co.