Models that match your story
Fast free tiers for iteration, pro models when you are ready to burn credits on quality. One router, consistent error shapes.
OpenAI-style drop-in · one API for local & cloud · without the vendor circus
Run AI in production without the headaches.
Drop-in AI that works like OpenAI—often cheaper—with the same clients you already use. One key routes free local models and paid cloud tiers: chat, generate, embed, and web search, plus limits your team (and finance) can live with. Try the live playground on free models—no signup or card.
Reason to switch: familiar API, friendlier economics
- Playground · ship a prompt before signup
- Key rotation & usage logs
- Queue visibility
- Free + paid model tiers
-
Same verbs as
/v1
One API for local + cloud models
Prototype on free tiers, flip to faster cloud models when you are ready—same contracts, same dashboard, no second platform to babysit.
Docs you can skim
Authentication, streaming semantics, and limits in one place—no PDF archaeology.
Keys on your terms
Issue, rotate, and audit API keys from the dashboard without downtime.
Web search as a plain POST
Grounded retrieval is not a separate “enterprise module.” It is POST /api/v1/web-search—same auth, same mental model as chat.
Plan-gated quotas keep costs predictable.
Streaming done right
NDJSON for native routes, SSE where OpenAI clients expect it—pick what fits your stack.
Billing-aware limits
Daily and weekly caps on paid models, searchable web quotas by tier—so surprise bills do not become your brand.
OpenAI-style calls, or native REST—your pick.
Point your OpenAI client at our /v1 base and ship today; use native JSON when you want full control. Same models, local or cloud, one bill to think about.
-
X-API-Keyeverywhere on REST - Bearer option on compatible routes
- Embeddings, chat, generate, queue status
Python
OpenAI SDK
from openai import OpenAI
client = OpenAI(
base_url="https://api.trinix.gg/v1",
api_key=os.environ["TRINIX_API_KEY"],
)
r = client.chat.completions.create(
model="trinix-chat",
messages=[{"role": "user", "content": "Ship faster."}],
)Pricing that scales with seriousness
Not ready to register? Use the playground to test free models. Start free, upgrade when you need cloud models, higher prompts, or web search.
Free
Prototype & learn
$0/mo
- Prompt cap · 4k chars
- 100 requests/min · 1 MB body
- GPU queue · max 50 waiting, excess → 429
- Free models · no daily quota
- Web search · not included
Pro
For shipping teams
$10/mo
- Prompt cap · 100k chars
- 100 requests/min · 1 MB body
- GPU queue · max 50 waiting, excess → 429
- Paid models · 500/day, 3,500/week
- Web search · 10/day
Enterprise
For heavy workloads
$199/mo
- Prompt cap · 200k chars
- 100 requests/min · 1 MB body
- GPU queue · max 50 waiting, excess → 429
- Paid models · 1,000/day, 5,000/week
- Web search · 50/day · up to ~100 on contract
Test in the browser, then wire up the API.
Open the playground with no account—try local models and streaming—then sign up when you want API keys, saved chats, and cloud tiers. Already decided? Grab a key and hit /api/v1/models in under a minute.