GLM
Agentic coding and long-horizon tasks. The launch workhorse.
R9SLeading open modelsone API away
Use GLM, Kimi, DeepSeek, Qwen, and MiniMax through one OpenAI-compatible endpoint, with usage, cost, and latency visibility on every call.
from openai import OpenAI
client = OpenAI(
base_url="https://gw.r9s.ai/v1",
api_key=os.environ["R9S_API_KEY"],
)
resp = client.chat.completions.create(
model="glm",
messages=[{"role": "user", "content": "Refactor this module."}],
)import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://gw.r9s.ai/v1",
apiKey: process.env.R9S_API_KEY,
});
const resp = await client.chat.completions.create({
model: "glm",
messages: [{ role: "user", content: "Refactor this module." }],
});curl https://gw.r9s.ai/v1/chat/completions \
-H "Authorization: Bearer $R9S_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "glm",
"messages": [
{ "role": "user", "content": "Refactor this module." }
]
}'Every family. One key.
These open models now go toe-to-toe with frontier models for coding agents, long-context work, and research — at a fraction of the cost.
Kimi
Long-context retrieval, document and research workflows.
DeepSeek
High-volume reasoning and coding at low cost
Qwen
General reasoning, multilingual and vision workloads.
MiniMax
Text, voice and multimodal agent workloads.
Access and payment are the real blockers
Getting to these open models is rarely a model problem. The hard part is everything around the model.
One endpoint, every model
A single OpenAI-compatible API across every family. Point your existing SDK at R9S and switch models with one line.
Pay like any other dev tool
Top up with the PayPal account you already use. No special setup, no paperwork — you are running in minutes.
Tokens at cost during launch
For a limited time, top up and we waive the 5% service fee, so you pay exactly what the tokens cost. Routing, failover, and controls are included.
Ready for a team
Owner, admin, and member roles; per-member spend limits; and scoped API keys you can revoke in one click.
Run your first call in about
three minutes
Create an account
Sign up with email, GitHub or Google. No sales call, no wallet, no paperwork.
Top up with $5
Add prepaid credits. For a limited time, the 5% service fee is waived on top-ups.
Connect your Agent or Application
Create your first API key, then use the OpenAI-compatible endpoint to start routing model calls with minimal code changes.
Token cost Zero service fee
For a limited time, top up and R9S waives the 5% service fee.
Available now
Pay as you go
Prepaid credits. During launch, top-ups are billed at token cost. Spend what you load. Nothing recurring.
$5.00minimum top-up
0 service fee- Every model family on one key
- Cross-provider routing and automatic failover
- Team roles, spend limits, scoped keys
- Usage and cost breakdown by model, project and key
Talk to usEnterprise
For teams running production workloads at volume, with procurement, compliance and finance in the room.
Custom
- Volume pricing committed-use rates that improve with scale
- SLA options matched with a named support channel
- Metering and chargeback usage attributed by team, project and cost center
- Invoicing, PO and annual billing
- SSO, audit trails and data-handling review
- Dedicated capacity and onboarding support
Looking for per-model token prices? See the full model list
The questions everyone asks first
Which models can I use?+
GLM, Kimi, DeepSeek, Qwen and MiniMax at launch, with more on the way.
What do I need to sign up and pay?+
An email or social login and a $5 prepaid top-up. No recurring subscription is required.
Is there really no service fee?+
During launch, R9S waives the 5% service fee on top-ups, so you pay token cost. This is a limited-time offer.
Is it really OpenAI-compatible?+
Yes. Keep your existing OpenAI SDK and change the base URL, API key and model name.
What comes with Enterprise?+
Volume pricing, procurement support, SSO review, audit trails, reserved capacity and guided onboarding.