Models

One API. Our own models.

PlusAPI serves our own models through a single OpenAI-compatible endpoint. AEON is the first — more will join the same API and account, each with its own name.

FLAGSHIP

AEON

aeon-27b

A capable 27B model served through an OpenAI-compatible endpoint — fast, private, and drop-in. Change one line and ship.

Parameters	27B (dense)
Modality	Text · chat completions
Context window	up to 64K tokens (by plan: 8K / 16K / 32K / 64K)
API	OpenAI Chat Completions — non-stream & streaming (SSE)
Model ID	aeon-27b
Endpoint	https://plusapi.ai/v1

Own model Zero logs Continuous batching Drop-in OpenAI SDK

curl https://plusapi.ai/v1/chat/completions \
  -H "Authorization: Bearer sk-aeon-…" \
  -d '{"model":"aeon-27b","messages":[{"role":"user","content":"Hello"}]}'

Get a key Read the docs

Usage & limits

Any lawful use. Requests that violate our Acceptable Use Policy — for example, content sexualizing minors — are blocked and return 400.
Nothing kept. Prompts and outputs are never stored (Privacy). What you send stays yours.
OpenAI-compatible. Standard /v1/chat/completions, streaming, and usage token counts.

Performance & capacity

AEON runs on a self-managed multi-GPU fleet with continuous batching, so it holds up under concurrent load. Throughput and parallelism scale with your plan — each plan sets max parallel requests, requests-per-minute, and context window (full table in the docs). Need guaranteed capacity? A Dedicated box gives you a private instance with an uptime SLA.

More models are coming

We're expanding the catalog. New models will appear here under their own names and IDs on the same endpoint and the same key — your integration won't change. Want a specific open model hosted? Tell us.